EAGLE888 AI HUMAN

CONSOLIDATED TERMS & CONDITIONS,

PRIVACY POLICY AND DATA ETHICS FRAMEWORK

For AI Human Workforce, AI Consulting and AI Pandit (Anika)

Effective: 27 September 2026 | Consolidated Version 1.1

How this Consolidated Document operates

One checkout acceptance is intended to cover this integrated package. The Subscriber agrees to Part I — Terms and Conditions of Service, acknowledges Part II — Privacy Policy, and agrees to comply with any Subscriber obligations expressly stated in Part III — Data Ethics & Responsible AI Framework. Eagle888’s commitments in Parts II and III remain binding to the extent incorporated into the Services, the Terms, an Order Form, SOW, DPA or applicable law. Acceptance of this Consolidated Document is not a substitute for separate express consent where applicable law requires separate, informed, voluntary, current and specific consent, including certain collections of sensitive information, marketing communications, cookies or similar technologies. Any such consent will be obtained separately when required. Order of precedence. If there is an inconsistency: (a) mandatory law applies to the extent it cannot lawfully be excluded; (b) an applicable DPA, SOW or Order Form prevails for the specific processing or assignment it expressly governs; (c) Part I governs the contractual Service relationship; (d) Part II governs privacy transparency and personal-information handling; and (e) Part III governs responsible-AI operating standards and applicable Subscriber conduct. No part of this Consolidated Document is interpreted to remove a non-excludable statutory right.

AI Consulting operational activation. For AI Consulting, checkout acceptance and successful payment authorise Eagle888 to verify the payment server-side, allocate the purchased entitlement or credits, create or associate the Subscriber tenant and engagement records required for the service, and make the authorised assignment workflow available. A payment-success screen is not, by itself, the technical record of entitlement. If fulfilment is delayed by a technical issue, Eagle888 may reconcile the existing paid transaction after verification and should not require a duplicate payment solely because the original fulfilment was delayed.

PART I — TERMS AND CONDITIONS OF SERVICE

IMPORTANT NOTICE — PLEASE READ BEFORE ACCEPTING

Eagle888 AI Human personas (AI Human Workforce personnel, AI Consultants, and Anika — AI Pandit) are artificial-intelligence software systems and digital service-delivery agents. They are not real human beings, natural persons, directors, officers, or employees, regardless of any functional title, persona, image, avatar, or manner of communication. Artificial intelligence can make mistakes and may generate output that is incomplete, inaccurate, or inappropriate. Before any AI-generated output is relied upon, implemented, or transmitted externally, it must be reviewed and approved by a responsible human person.

PART A — WHO WE ARE

Service Provider: Eagle888 AI Human Pty Ltd (ABN 52 676 753 423; ACN 676 753 423) of 40 Koola Avenue, East Killara NSW 2071, Australia (“Provider”, “Eagle888”, “we”, “us”). Intellectual Property / Patent Owner: Eagle888 Pty Ltd (ABN 87 624 709 486; ACN 624 709 486), a related entity, is the owner of the patents, trademarks, software, and other intellectual property in the AI Human platform and the AI Human personas (including “Anika — AI Pandit”). The Provider is authorised by the patent owner to deliver the Services under these Terms. In these Terms, “Eagle888 Group” means the patent owner and the Provider (and their related bodies corporate). Nothing in these Terms grants the Subscriber any right in the patent owner’s intellectual property beyond the limited licence in Part J.

PART B — THE THREE SERVICES

Eagle888 provides three distinct services (together, the Services, and each a Service Vertical):

# Service Vertical What it is Contractual basis

1 AI Human Workforce Subscription to configured AI-powered digital workers (“AI Humans”) for defined professional and operational roles, delivered as a recurring SaaS subscription with credit-based usage. These Terms + Schedule 1 2 AI Consulting (AI Human Consultants) Defined consulting assignments in technology, innovation, growth, sustainability, finance, governance, risk, and related professional disciplines, delivered by a digital AI Consultant (an “AI Human Consultant”). These Terms + Schedule 2 + an Proposal / Order Form / Statement of Work (“SOW”) 3 AI Pandit (Anika) AI-assisted spiritual, cultural, and wellbeing information services, including digital reports and remotely guided home puja services. These Terms + Schedule 3 Each Service Vertical has vertical-specific provisions in its Schedule. The general terms in Parts C-N apply to all Service Verticals except where a Schedule expressly modifies them. If there is any conflict, the vertical Schedule prevails over these general terms for that Service Vertical.

PART C — DEFINITIONS

Term Meaning

Provider Eagle888 AI Human Pty Ltd (ABN 52 676 753 423; ACN 676 753 423). Patent Owner Eagle888 Pty Ltd (ABN 87 624 709 486; ACN 624 709 486), owner of the AI Human platform IP. Subscriber The person or entity that creates an account and subscribes to a Service Vertical. AI Human A virtual, software-based digital worker instance provisioned under the AI Human Workforce Vertical (Schedule 1). AI Consultant A virtual, software-based digital consulting persona provisioned under the AI Consulting Vertical (Schedule 2). AI Pandit (Anika) The virtual, software-based digital spiritual/cultural/wellbeing information persona provided under Schedule 3. AI Human persona Collectively, any AI Human, AI Consultant or AI Pandit (Anika) instance. Services The Eagle888 AI Human cloud platform and the three Service Verticals, including all features, storage, and AI Backbone access. Subscription Tier The recurring pricing tier applicable to a Service Vertical, as published in the Pricing Schedule from time to time. Credits Service-consumption units. Credits are a metering mechanism and are not a guarantee of labour hours, output volume or a particular outcome. Business Hours The operating hours configured by the Subscriber during onboarding. Reporting Manager The human person designated by the Subscriber under the AI Human Workforce Vertical to review and approve AI Human outputs. Principal The human person designated by the client under the AI Consulting Vertical to review and approve AI Consultant outputs. AI Backbone Third-party AI services accessed by the AI Human persona (including OpenAI, Gemini, Perplexity, legal AI, and others) as determined by the Service Vertical and tier. Desk The AI Human persona’s persistent cloud workspace within Eagle888 or, where authorised, within the client’s workspace. Hot Desk The Subscriber’s shared cloud folder accessible to authorised AI Human personas within that subscription. Specialist Folder The AI Human persona’s role-specific work product folder. Deliverable Any output, document, report, communication or other work product generated by an AI Human persona.

Term Meaning

Overflow Folder An optional folder on the Subscriber’s own system for receiving working files or Deliverables. Pricing Schedule The then-current credit-based published pricing for the Services, which may be updated from time to time. Responsible Person The Reporting Manager, Principal, or other authorised human person required to review and approve an output, as applicable to the relevant Service Vertical.

PART D — ACCOUNT, ELIGIBILITY AND SUBSCRIPTION

D.1 Account creation. The Subscriber creates an account using Apple ID, Google ID, or an email address.

D.2 Accurate information. The Subscriber must provide accurate, current, and complete information and

update it promptly.

D.3 Credentials. The Subscriber is responsible for maintaining the confidentiality of its account credentials

and for all activity occurring under its account.

D.4 One account per entity. One account per legal entity. Sharing an account between unrelated

organisations is prohibited.

D.5 Eligibility. The Subscriber must be at least 18 years of age and have legal capacity to enter into a

binding agreement.

D.6 Vertical selection. The Subscriber may subscribe to one or more Service Verticals. Each Service

Vertical is contracted separately and may be cancelled independently, subject to Parts F and I.

PART E — FEES, PAYMENT AND CREDITS

E.1 Fees. Fees for each Service Vertical are set out in the Pricing Schedule (and, for AI Consulting, in the

applicable SOW), are quoted in Australian Dollars (AUD), and are debited in accordance with the Subscriber’s selected billing cycle.

E.2 GST, local taxes and payment-processor tax calculation. All prices and fees published or quoted by the

Provider are exclusive of GST and all other taxes, duties, levies and governmental charges, unless expressly stated otherwise. The Subscriber must pay any GST, VAT, sales tax, use tax, digital-services tax or other indirect tax legally applicable to the transaction in the Subscriber’s jurisdiction. Where Stripe Tax or another Stripe tax-calculation function is configured and available, applicable taxes may be calculated and collected by Stripe at checkout or invoicing based on the transaction, the Subscriber’s billing or service location, product tax classification, and the Provider’s tax registrations and settings. If applicable tax is not collected at checkout or invoicing, the Subscriber remains responsible for any tax that lawfully falls on the Subscriber. Currency-conversion, card, bank or intermediary fees charged to the Subscriber are the Subscriber’s responsibility.

E.3 Payment methods. Credit card, debit card, Stripe, PayPal, payment ID, or bank-account direct debit.

All payments are processed through PCI DSS-compliant payment gateways.

E.3A Payment verification and service activation. Payment processing and service activation are separate technical steps. For a paid Service, Eagle888 activates credits, entitlement, onboarding or an assignment only after server-side verification of the applicable payment or subscription event. A checkout confirmation page or payment receipt does not by itself prove that downstream fulfilment has completed. Where a verified payment has succeeded but fulfilment is delayed, the Provider may reconcile the existing transaction and restore the purchased entitlement without requiring the Subscriber to pay again for the same purchase. The Provider may use payment, Checkout Session, event or other transaction identifiers to prevent duplicate fulfilment.

E.4 Credits. Credits are consumed against tasks and Deliverables as described in the Pricing Schedule or

SOW. Credits do not expire during an active subscription (unless the Pricing Schedule states otherwise) and are not redeemable for cash. Credits may get used up during an active subscription due to use and complexity of the service required.

E.5 Non-refundability. All subscription fees and prepaid credits are non-refundable to the maximum extent

permitted by law, but nothing in this clause limits any non-excludable statutory right to a remedy or refund under the Australian Consumer Law or equivalent consumer-protection law applicable to the Subscriber.

E.6 Cancellation — use until end of paid period. If a Subscriber cancels a subscription, access to that

Service Vertical continues until the end of the then-current paid period. No pro-rated refunds are payable for unused credits and time, except as required by applicable law.

E.7 Upgrades and downgrades. Tier upgrades take effect immediately (the Subscriber is charged the pro-rated difference for the remainder of the current billing cycle). Downgrades take effect at the start of the

next billing cycle, at which point features exceeding the new tier’s limits are disabled.

E.8 Failed payments. If a payment fails, the Provider may retry the payment and, if payment remains

unsuccessful after reasonable attempts, may suspend or terminate the account in accordance with Part I.

PART F — SERVICE DESCRIPTION AND OPERATION

F.1 Platform and authorised integrations. The Provider operates the AI Human cloud platform from the

Eagle888 Group’s cloud infrastructure. The Services may connect to third-party applications, cloud services, websites, APIs, repositories, payment systems, accounting systems, productivity tools, or other environments only where the Subscriber expressly authorises that connection. The Provider will not intentionally access the Subscriber’s internal network, firewall, device, account, or system beyond the scope of the permissions granted by the Subscriber or otherwise permitted by law. The Subscriber is responsible for configuring and revoking third-party permissions made available to the Services.

F.2 Workspaces. Each AI Human persona is allocated a digital Desk, a dedicated email address, a core

folder, and a Specialist Folder. Each Subscriber is allocated a Hot digital Desk accessible to all AI Human personas within that subscription.

F.3 Communication. AI Human personas communicate via email and shared cloud links.

F.4 Business hours. AI Human personas operate during the Subscriber’s configured Business Hours; AI

Backbone and asynchronous processing may nonetheless generate output outside those hours.

F.5 AI Backbone. AI Human personas access third-party AI Backbone services as determined by the

Service Vertical and Subscription Tier. The Provider does not guarantee the continued availability, quality, or accuracy of any third-party AI Backbone.

F.5A AI Backbone dependency and change. The Services depend in part on third-party AI Backbone

providers and related third-party services that the Provider does not own or control. Their models, APIs, features, safety controls, usage limits, availability, latency, capacity, pricing, policies, data-handling settings, processing locations, retention practices and terms may change, be restricted, suspended or discontinued. The Provider may replace, reroute or reconfigure an AI Backbone where reasonably necessary for security, continuity, quality, compliance, cost or availability. A material change that materially disadvantages a Subscriber will be handled in accordance with Part N and applicable law.

F.5B No inference of correctness. Use of one or more AI Backbones, or agreement between multiple

models, does not establish that an output is accurate, complete, current or suitable. Human review obligations in Part H continue to apply.

F.6 Storage and notifications. Storage is limited to the allocation in the applicable tier, published from

time to time. The Provider will notify the Subscriber when usage reaches 100% of the allocation.

F.7 Overflow Folder. The Subscriber may create an Overflow Folder on its own system. The AI Human

persona will send working and Deliverables to the Overflow Folder by email or shared link upon request.

F.8 Status of AI Human personas. Every AI Human persona is a digital, virtual, software-based

contracted service agent operated by the Provider. It is not a natural person and not an employee, director, officer, agent at law, or contractor of the Subscriber for employment, taxation, workers’ compensation, or superannuation purposes.

PART G — DATA OWNERSHIP, SECURITY AND PRIVACY

G.1 Ownership of Subscriber data. All data uploaded by the Subscriber and all Deliverables generated

on the Subscriber’s behalf remain the property of the Subscriber, subject to Part J (Intellectual Property) and any statutory limitation on the copyrightability of fully AI-generated material.

G.2 Storage and isolation. Data is stored within the Eagle888 Group cloud environment and is logically

isolated from other Subscribers’ data.

G.2A Subscriber and engagement access isolation. For AI Consulting and other supported workflows, the Service may enforce access through a server-verified context that associates an authenticated user or AI Human with a Subscriber tenant, an engagement or project, a Principal or other authorised identity, and a permitted role. Email addresses may be retained as account or notification metadata but are not intended to be the sole security boundary. An authorised Principal or AI Human must not be given access to another Subscriber or unrelated engagement merely because that party knows an identifier. Persistent storage, memory, retrieval, connected applications and provider context must be configured consistently with the applicable access boundary before they are used for Subscriber data.

G.3 No training on Subscriber data. The Provider will not use Subscriber data to train AI models that are

shared with other subscribers, except in anonymised, aggregated form for system-wide performance improvement, and only where the Subscriber has consented. The Provider will not use Subscriber data to train or refine models that the Provider, or any third party, makes available to others in a form that identifies the Subscriber or its confidential information.

G.4 Backups and resilience. The Provider maintains backup and disaster-recovery measures appropriate

to the relevant Service and hosting configuration. Where backups are maintained, the Provider applies security controls appropriate to the sensitivity of the data and the technical capability of the hosting service. The Subscriber remains responsible for maintaining its own independent copies of critical data and Deliverables.

G.5 Records-management folder. The Subscriber may create a work folder in its own records system into

which an AI Human persona saves work. Access to that folder is immediately suspended on termination of the subscription (see Part I).

G.6 Security safeguards. The Provider will maintain reasonable technical and organisational safeguards

appropriate to the nature of the Services and the sensitivity of the information processed. Safeguards may include encryption in transit and at rest, role-based access control, audit logging, identity and access controls, monitoring, data minimisation, masking or redaction, and controls designed to reduce prompt-injection and other AI-specific risks. Specific controls may vary by infrastructure component, Service Vertical, integration, and third-party sub-processor.

G.6A Cybersecurity and technological limitations. Eagle888’s ability to protect cybersecurity, privacy, confidentiality and communications is limited by the technological capabilities and safeguards reasonably available in the systems, networks, AI services and third-party providers used to deliver the Services. Eagle888 will take reasonable technical and organisational steps appropriate to the information and risks, within those capabilities, and will review material limitations as technology and provider provisions change. No technology can guarantee absolute security, uninterrupted communications or prevention of every unauthorised access event. These limitations do not excuse Eagle888 from its obligations under applicable privacy, security or consumer law, or any expressly agreed contractual safeguard.

G.7 Data-breach response and notification. The Provider will investigate suspected data breaches

affecting Subscriber data and will notify the Subscriber without undue delay where notification is required by applicable law or is reasonably necessary for the Subscriber to meet its own legal obligations. Where the Australian Notifiable Data Breaches scheme applies, the Provider will act in accordance with the Privacy Act 1988 (Cth) and applicable OAIC requirements. Different timeframes may apply under overseas privacy laws, including GDPR obligations where applicable.

G.8 Security review and testing. The Provider will periodically review and test security controls using

measures proportionate to the risks of the relevant systems and Services. Any published certification, audit, penetration-testing, or review claim applies only where the Provider has current evidence supporting that claim.

G.9 Privacy and cross-border laws. The Provider complies with the Australian Privacy Act 1988 (Cth) and

the Australian Privacy Principles. Where applicable, the Provider complies with the EU/UK General Data Protection Regulation and the California Consumer Privacy Act (CCPA/CPRA). Where the Subscriber is subject to the CCPA (or GDPR), the parties’ respective roles and obligations are set out in the applicable data-protection addendum in Schedule 4. The Subscriber acknowledges that the Provider may process personal information in Australia (and, where a third-party sub-processor is used, in the sub-processor’s jurisdiction) in accordance with applicable law.

PART H — AI LIMITATIONS, HUMAN OVERSIGHT AND INSURANCE

H.1 AI can make mistakes. The Subscriber acknowledges that each AI Human persona is an artificial-intelligence system that can generate output that is incomplete, inaccurate, or inappropriate, despite

the Provider’s safeguards. AI-generated output must not be assumed to be correct merely because it has been produced or delivered by an AI Human persona.

H.2 Informational and decision-support only. AI Human persona outputs are provided for informational

and decision-support purposes only and do not constitute legal, financial, tax, medical, psychological, religious, spiritual, or other regulated professional advice, and are not a substitute for advice from a real human, educated with formal qualifications, experience, qualified, licensed professional.

H.3 Mandatory human review. The Subscriber must designate a Responsible Person for each AI

Human persona. All important or externally-facing outputs must be reviewed, verified, and approved by that Responsible Person (or another authorised human) before being used, executed, published, acted upon, or relied upon. Without limiting this clause, the Subscriber must not rely on AI Human persona outputs as the sole basis for:

• any use, external communications (including marketing, investor, regulatory, or client

communications);

• financial, tax, employment, safety, health, or legal decisions; or

• any decision that could reasonably pose a risk to people, property, or reputation.

H.4 No provider indemnity, liability for failure to review. The Provider is not responsible for any loss,

immediate or consequentially arising from the Subscriber’s failure to implement appropriate human review and approval, subject to any non-excludable consumer-protection rights.

H.5 Insurance. During the term, the Subscriber must maintain, at its own cost, insurance appropriate to its

organisation, size, and industry to cover its use of the Services, including (where applicable) public-liability and professional-indemnity (errors and omissions) insurance that responds to claims arising from the Subscriber’s use of the Services and AI Human persona activities performed under the Subscriber’s instructions. On reasonable request, the Subscriber will provide certificates of currency.

H.6 Provider insurance. Where the Provider maintains public-liability, professional-indemnity, cyber, or

other insurance, that insurance does not create rights in favour of the Subscriber beyond the applicable policy and law. Nothing in this Part H limits or expands the Provider’s liability beyond Part L.

PART I — TERMINATION, UNSUBSCRIPTION AND DATA

I.1 Unsubscription. The Subscriber may unsubscribe from any Service Vertical at any time through the

subscriber portal.

I.2 Effect. Cancellation takes effect at the end of the then-current paid period for that Service Vertical

month, during which the Subscriber may continue to use that Service Vertical.

I.3 Data export offer. Upon initiating unsubscription, the Subscriber will be offered a data-export option and

will have up to the end of the subscription period to download stored data.

I.4 Election. The Subscriber may elect to (a) download all data held in its Hot Desk and Specialist Folders,

or (b) confirm that all data should be permanently destroyed.

I.5 Destruction and certificate. After the data-export period, the Provider will permanently destroy all

Subscriber data and deactivate the AI Human persona instance(s) and associated email address(es). A Data Destruction Certificate will be issued, only if required by the law.

I.6 Records-management folder. On termination, any access by an AI Human persona to a folder the

Subscriber created in its records-management system is suspended immediately and automatically.

I.7 Termination by the Provider. The Provider may terminate a Subscriber’s account for a material breach

of these Terms on fourteen (14) days’ written notice, or immediately for a security breach, unlawful use, or a breach of Part K (Unacceptable Breach) that poses an ongoing risk.

I.8 Survival. Parts G (Data), H (AI Limitations, Oversight and Insurance), I.5–I.6 (Data destruction), Part J

(Intellectual Property), Part L (Limitation of Liability), Part M (Governing Law and Disputes), and Part N (General) survive termination.

PART J — INTELLECTUAL PROPERTY

J.1 Platform IP. The AI Human platform (including software, algorithms, models, system architecture, and

personas) and all patents, copyright, and other IP in it are the property of the Patent Owner (Eagle888 Pty Ltd) or its licensors.

J.2 Subscriber ownership of content and Deliverables. The Subscriber retains ownership of all content

and data it provides. To the maximum extent permitted by law, the Provider assigns to the Subscriber all of its right, title, and interest, if any, in and to the Deliverables generated for the Subscriber. The Subscriber acknowledges that:

• fully AI-generated output, without material human selection, arrangement, or modification, may not

be independently subject to copyright protection in certain jurisdictions (including the United States); and

• similar, or identical, output may be generated for other subscribers, and the ownership granted in

this clause does not confer exclusivity over any such output.

J.3 Licence to Provider. The Subscriber grants the Provider a non-exclusive, limited licence to use

Subscriber data solely for the purpose of providing, operating, and improving the Services, and as otherwise required by law.

J.4 Licence to Subscriber. The Provider grants the Subscriber a non-exclusive, non-transferable, non-sublicensable licence to use the Services during the subscription term for the Subscriber’s internal business

purposes.

J.5 No conveyance of patent rights. Nothing in these Terms conveys or licenses any patent, trade mark,

or other IP of the Eagle888 Group to the Subscriber beyond the licence in clause J.4.

PART K — UNACCEPTABLE BREACH

K.1 Unacceptable Breach. The Subscriber must not use, direct, authorise or permit the Services to:

• engage in unlawful, fraudulent, deceptive, defamatory, discriminatory, harassing, abusive or

malicious conduct;

• create, facilitate, distribute or promote child sexual abuse material, sexual exploitation of minors,

grooming, trafficking, non-consensual intimate content, or other unlawful sexual exploitation;

• create or materially facilitate instructions, designs, code, substances, processes or operational

assistance intended to cause or materially enable physical injury to self or to other person or any real human, self-harm, suicide, violence, abuse, terrorism or other serious harm;

• generate unlawful pornography or sexually explicit material, or harmful sexual or gender-related

harassment, abuse, exploitation, violence or unlawful discrimination;

• infringe intellectual-property, privacy, publicity, confidentiality or other legal rights, including by

reproducing or impersonating a real person’s voice, image or likeness without lawful authority;

• bypass, defeat or interfere with security, cyber-security, access-control, safety, monitoring, rate-limit

or usage-control mechanisms;

• reverse engineer, decompile, disassemble, scrape, probe or attempt to obtain source code, system

prompts, model weights, credentials, confidential architecture or non-public technical information except to the extent a right cannot lawfully be excluded;

• use the Services primarily to build, train or operate a competing AI Human platform by copying

protected platform functionality, confidential information or proprietary architecture;

• falsely represent an AI Human persona as a natural person, director, employee, licensed

professional or regulated adviser;

• make automated calls, messages, marketing communications or consumer contacts in breach of

applicable consent, disclosure, anti-spam, telemarketing or similar laws; or

• use the Services in a manner that creates a material and unreasonable risk to persons, property,

systems, the Provider, third parties or the integrity of the Services.

K.2 Suspension. The Provider may suspend access for a breach of this Part K with immediate effect,

pending investigation.

K.3 AI backbones. The Subscriber’s use of the Services is also subject to the acceptable-use policies of

the third-party AI Backbone providers, to the extent applicable.

PART L — LIMITATION OF LIABILITY

L.1 No warranties. To the maximum extent permitted by applicable law, the Services (including all AI

Human persona outputs) are provided “as is” and “as available”. The Provider does not warrant that any AI Human persona output will be accurate, complete, current, or error-free, and disclaims all implied warranties, including merchantability, fitness for a particular purpose, and non-infringement.

L.2 Exclusion of indirect damages. To the maximum extent permitted by applicable law, neither party will

be liable for any indirect, incidental, special, consequential, exemplary, or punitive damages, or for any loss of profits, revenue, savings, goodwill, data, or business interruption, arising out of or relating to the Services or these Terms, even if advised of the possibility of such damages.

L.3 Cap. Subject to clause L.4, each party’s total aggregate liability arising out of or relating to the Services

and these Terms in any period will not exceed the greater of (a) the fees actually paid by the Subscriber for the affected Service Vertical(s) during the month of the subscription period, or (b) AUD $100 (the “Liability Cap”), whichever is more.

L.4 Exceptions to the cap. The Liability Cap does not apply to:

• the Subscriber’s payment obligations;

• either party’s liability for fraud or fraudulent misrepresentation;

• either party’s liability for death or personal injury caused by negligence;

• either party’s liability for wilful misconduct or gross negligence;

• the Provider’s infringement of the Subscriber’s ownership rights in its data or Deliverables (clause

J.2), to the extent caused by the Provider’s material breach;

• any liability that cannot be excluded or limited under applicable law.

L.5 Mandatory rights – savings. Nothing in these Terms excludes, restricts or modifies any right,

guarantee, remedy or liability that cannot lawfully be excluded, restricted or modified. Any such mandatory right applies only to the extent required by law and does not otherwise vary the parties’ express contractual choice of New South Wales law and forum under Part M.

L.6 AI Backbone and third-party service limitations. The Services rely on AI Backbones and other third-party providers outside the Provider’s direct control. To the maximum extent permitted by law, the Provider

is not liable for loss, delay, interruption, degradation or error caused solely by an act, omission, outage, model behaviour, policy change, security event or discontinuance of a third-party provider where the Provider has exercised reasonable care and the matter is outside the Provider’s reasonable control. This clause does not exclude or limit liability to the extent caused by the Provider’s own breach, negligence, wilful misconduct, failure to apply reasonable provider governance, or any liability, guarantee, remedy or right that cannot lawfully be excluded or limited.

PART M — GOVERNING LAW AND DISPUTES

M.1 Exclusive governing law. These Terms, every Subscription, Order Form, SOW, transaction, non-contractual obligation, and any dispute or claim arising out of or in connection with the Services or the

parties’ relationship are governed exclusively by the laws in force in New South Wales, Australia, without regard to conflict-of-laws principles. To the fullest extent permitted by law, no foreign governing law is incorporated into or substituted for these Terms merely because a Subscriber is located outside Australia.

M.2 Exclusive jurisdiction. To the fullest extent permitted by law, each party irrevocably submits to the

exclusive jurisdiction of the courts of New South Wales, Australia, and the Commonwealth courts of Australia sitting in New South Wales, and waives any objection based on venue, inconvenient forum or similar grounds. If a mandatory law that cannot legally be excluded requires proceedings to be heard elsewhere, that mandatory requirement applies only to the minimum extent legally required and does not otherwise alter the parties’ agreement that New South Wales is the exclusive contractual forum.

M.3 Negotiation. Before commencing court proceedings, the parties will attempt to resolve disputes

through good-faith negotiation for a period of thirty (30) days.

M.4 Mediation. If unresolved, either party may refer the dispute to mediation administered by the Australian

Disputes Centre.

M.5 International Subscribers. A Subscriber’s location outside Australia does not change the governing

law or contractual forum in clauses M.1 and M.2. Mandatory local consumer, privacy, tax or regulatory requirements that apply by operation of law are preserved only to the extent they cannot lawfully be excluded. Compliance with such requirements does not constitute the Provider’s submission to a foreign court or foreign governing law except where that result is itself legally mandatory.

PART N — GENERAL

N.1 Amendments. The Provider may amend these Terms on at least thirty (30) days’ written notice where

the change is material. Changes required by law, security, fraud prevention, or a third-party platform dependency may take effect sooner where reasonably necessary. Material changes will not apply retroactively. If a material change materially disadvantages the Subscriber, the Subscriber may cancel the affected Service Vertical before the change takes effect, without an early-termination penalty; any non-refundable amounts remain subject to applicable consumer law.

N.2 Assignment. The Subscriber may not assign its rights without the Provider’s prior written consent. The

Provider may assign the agreement to a successor entity or an affiliate within the Eagle888 Group.

N.3 Force majeure. Neither party is liable for delays caused by events beyond reasonable control,

including natural disasters, pandemics, government actions, and internet and system outages of the Providers, related third parties and Backbones.

N.4 Severability. If any provision is held invalid or unenforceable, the remaining provisions continue in full

force.

N.5 Entire agreement. These Terms, together with the Pricing Schedule, any Order Form, and any SOW,

constitute the entire agreement between the parties.

N.6 Notices. All notices must be in writing and sent to the email address associated with the account

(Subscriber) or to aihuman@eagle888.co (Provider).

SCHEDULE 1 — AI HUMAN WORKFORCE

S1.1 Scope. The AI Human Workforce Vertical provides configured AI-powered digital workers for defined

professional and operational roles, under a recurring subscription with credit-based usage.

S1.2 Subscription tiers and credits. The Subscriber selects a Subscription Tier from the Pricing Schedule

(for example, an evaluation tier, Standard, Professional, Advanced, or Enterprise), each of which includes a defined number of monthly credits. Credits are service-consumption units and are not a guarantee of labour hours, output volume or any desired outcome.

S1.3 Roles and governance. The Subscriber must designate a Reporting Manager for each AI Human.

The Reporting Manager (or another authorised human) is the Responsible Person for the purposes of Part H and must review and approve AI Human outputs before use.

S1.4 Configuration and activation. Following acceptance of these Terms, the Subscriber provides its

organisation details, AI Human role, Reporting Manager, permitted functions, Business Hours, and operating requirements; the Provider then provisions the AI Human instance, identity, and Desk.

S1.5 Workflow. Authorised users assign tasks, documents, and instructions; the AI Human processes

them through the applicable quality, risk, and human-validation controls of the Reporting Manager for delivery of the output. Authorised users have the flexibility of making changes to the assign tasks, documents, and instructions.

S1.6 Evaluation tier. Any paid evaluation tier is a credit limited evaluation and does not automatically

convert into a continuing subscription, unless the Subscriber elects to resubscribe and/or upgrade.

SCHEDULE 2 — AI CONSULTING (AI HUMAN CONSULTANTS)

S2.1 Scope. The AI Consulting Vertical provides AI Human Consultants for defined consulting assignments

in technology, innovation, growth, sustainability, finance, governance, risk, operations, digital transformation, research, business analysis, and related professional disciplines. An AI Human Consultant is an artificial-intelligence software service and not a natural person, employee, director, officer, partner, fiduciary, agent at law, or licensed professional.

S2.2 Assignment and SOW. Each assignment is delivered as per the Proposal, Order Form, or Statement

of Work (“SOW”) that specifies the scope, assumptions, Deliverables, credits, timetable, client dependencies, acceptance criteria, authorised systems or data access, and any special governance controls provided by the Subscriber. The Principal or Subscriber may amend the SOW, Order Form, scope, assumptions, Deliverables or timetable, including additional credits, subject to the Provider’s written confirmation.

S2.2A Purchase, entitlement and onboarding. Where AI Consulting credits or an assignment entitlement are purchased through Stripe or another approved checkout, access to the assignment workflow is activated after server-side verification of the successful payment event and the applicable entitlement. The service process may create or associate a Subscriber tenant, an engagement identifier, credits and an authorised Principal before presenting the assignment start or onboarding step. A technical delay in fulfilment does not require a second payment for the same verified entitlement. Eagle888 may reconcile the original paid transaction and may use Stripe or other transaction identifiers to avoid duplicate crediting or duplicate fulfilment.

S2.3 Principal and human accountability. The Subscriber must designate a Principal who is a

responsible human person with authority to supervise the assignment. The Principal is the Responsible Person for Part H and remains accountable for reviewing, validating, approving, rejecting, and authorising use of AI Consultant outputs. The AI Consultant has NO authority to bind the Subscriber or the Provider, enter contracts, approve expenditure, make regulated decisions, sign documents, submit regulatory filings, or communicate externally in the Subscriber’s name.

S2.3A Confidential assignment boundary. Each AI Consulting assignment is associated with the relevant Subscriber and engagement or project boundary. Access is limited to authenticated and authorised Principals, AI Human personas and Provider roles for that engagement. The service may use non-guessable tenant and engagement identifiers, signed access context, role permissions and audit records to enforce that boundary. A customer email address may be retained as authorised account metadata but is not intended to be the tenant-isolation key.

S2.4 Professional-services disclaimer. AI Consultant output is provided for research, drafting, analysis,

workflow support, and decision-support purposes only and is not professional advice. Where an assignment touches a regulated or licensed discipline, including legal, financial, tax, accounting, engineering, medical, psychological, employment, safety, investment, or other regulated advice, the Subscriber MUST obtain review or advice from an appropriately qualified and licensed human professional before relying on or implementing the output.

S2.5 Client information and dependencies. The Subscriber is responsible for providing timely, accurate,

lawful, and sufficiently complete instructions, data, documents, access permissions, and decisions needed for the assignment. The Provider and AI Consultant may rely on information supplied by or on behalf of the Subscriber unless the SOW requires independent verification, it must be obtained by the Subscriber. Delays, omissions, inaccessible systems, inaccurate data, changed assumptions, real human reviews by the Subscriber and the Authorized Persons and the Principal or delayed approvals may affect the timetable, credits, fees, and quality of Deliverables.

S2.6 Authorised access and integrations. Any access by an AI Consultant to the Subscriber’s website,

cloud environment, repositories, email, accounting, payment, CRM, file storage, analytics, or other systems must be expressly authorised and limited to the permissions reasonably required for the SOW. The Subscriber must not provide credentials or permissions that exceed what is reasonably necessary and must promptly revoke access when no longer required. High-risk actions must require human confirmation or may be excluded from autonomous execution.

S2.7 Change control. Work outside the initial proposed scope, including material changes to objectives,

Deliverables, integrations, data volumes, jurisdictions, or deadlines, may require a written change to the SOW, additional credits, or a revised timetable. The Provider will not treat a material scope change as accepted solely because an AI Consultant discussed or drafted it. Any change must always be verified and approved by the Subscriber or the Principal or the Authorised person to the Subscription.

S2.8 Deliverable review and acceptance. Unless the SOW states a different period, the Subscriber

should review each material Deliverable promptly and notify the Provider of any material non-conformity with the agreed acceptance criteria. A Deliverable is accepted when the Principal expressly approves it, uses it for its intended business purpose after a reasonable opportunity for review, or otherwise confirms acceptance. Acceptance does not waive non-excludable rights under applicable law.

S2.8A Independent AI quality review and Principal review. Where configured, an AI Consultant draft may pass through a separate AI quality-review stage before it is released to the Principal. A draft that is marked for revision is withheld from Principal release until the applicable workflow permits release. A draft that passes the independent review may be released to the Principal as a draft for human review; the detailed internal review may remain internal. Passing an AI review does not make the Deliverable final and does not replace the Principal’s human responsibility under S2.3. Unless an SOW or applicable workflow states otherwise, the standard Principal review window may be set at fourteen (14) days from the time the draft is released for Principal review. The Principal may approve, reject or request revision in accordance with the assignment workflow and these Terms.

S2.9 Confidentiality. Each party must protect the other party’s confidential information using reasonable

care and use it only for the assignment or as otherwise permitted by these Terms, the SOW, or law. This obligation does not apply to information that is public other than through breach, already lawfully known, independently developed, or lawfully obtained from a third party without confidentiality restriction.

S2.10 AI transparency. The Subscriber acknowledges, accepts and agrees that Deliverables and

interactions under this Vertical are generated or assisted by AI. Where the AI Consultant communicates with third parties or contributes to a decision affecting an individual, the Subscriber is solely responsible for any disclosure, notice, consent, contestability, recordkeeping, or human-review obligation that applies to the Subscriber’s use only.

S2.11 No guaranteed outcome. The Provider does not guarantee any commercial, financial, regulatory,

legal, technical, sustainability, fundraising, growth, or other outcome from an AI Consulting engagement. Credits are charged for usage under the applicable Subscription or SOW and not for achieving any particular result or outcome.

SCHEDULE 3 — AI PANDIT (ANIKA)

S3.1 Scope. The AI Pandit Vertical provides AI-assisted spiritual, cultural, and wellbeing information

services, including digital reports and remotely guided home puja services. It is provided for cultural, educational, informational and entertainment purposes only.

S3.2 Nature of the service. Anika — AI Pandit is a cultural- and information-based service. It is not a

religious authority, priest, astrologer, medical or psychological practitioner, or licensed professional of any kind. Services are provided for general cultural, educational, and informational purposes, entertainment, and are not a substitute for:

• outputs must not be relied upon as professional, medical, legal, financial, psychological or other

regulated advice;

• medical, mental-health, or psychological care from a licensed practitioner;

• the Provider does not intend to ridicule, denigrate or disparage any religion, faith, belief, mythology,

cultural tradition or spiritual practice;

• the Subscriber’s own religious, spiritual, or cultural discretion, or consultation with a religious or

cultural authority of the Subscriber’s choosing.

S3.3 Wellbeing disclaimer. Any wellbeing-related content is general information, not medical or

psychological advice. If the Subscriber or any person is experiencing a health, mental-health, or wellbeing concern, they should seek advice from a qualified health professional.

S3.4 Guided puja services. Remotely guided home puja services are conducted on a guided-information

basis only. The Subscriber is responsible for safe conduct of any physical activities, for compliance with local safety, fire, and building requirements, and for the suitability of the service to their own practices and beliefs.

S3.5 Responsible person. For the AI Pandit Vertical, the Responsible Person is the recipient of the report

or service (or, where applicable, another authorised human). Outputs must be reviewed, and are provided “as is”, subject to Part H.

S3.6 Refunds and consumer rights. To the extent the AI Pandit Vertical is supplied to a “consumer”

under the Australian Consumer Law, the Subscriber’s statutory guarantees and remedies under the Australian Consumer Law apply and are not excluded or limited by Part L.

SCHEDULE 4 — DATA-PROTECTION ADDENDUM (CCPA / GDPR)

S4.1 Application. This Schedule applies where the Subscriber is subject to the CCPA (California

Consumer Privacy Act / CPRA) or the GDPR/UK GDPR, and personal information/data of data subjects is processed through the Services.

S4.2 Roles. Where the Subscriber is a business/controller and the Provider processes personal information

on its behalf to provide the Services, the Provider acts as a service provider / processor, not a business/controller, in respect of that processing.

S4.3 Provider obligations. The Provider will (where acting as service provider/processor):

• process personal information only for the purpose of providing the Services and on the Subscriber’s

documented instructions;

• not “sell” or “share” (as defined in the CCPA) personal information, or otherwise retain, use, or

disclose it for any purpose other than the Services, except as permitted by law;

• comply with applicable deletion requests; and

• maintain reasonable and appropriate security safeguards.

S4.4 Sub-processors. The Provider uses third-party AI Backbone providers as sub-processors, for

commercial reasons the sub-processors are confidential and are only disclosed when required by law.

S4.5 Cross-border transfer. Where personal information is transferred across borders, the Provider will

comply with the applicable transfer mechanisms required by law.

SCHEDULE 5 — INTERNATIONAL SUBSCRIBERS

S5.1 Application. This Schedule applies to Subscribers located, incorporated or operating outside

Australia. It supplements the other provisions of these Terms and does not replace Part M.

S5.2 New South Wales governing law and forum. The Subscriber accepts, acknowledges and agrees

that the Services are contracted from the Provider in New South Wales, Australia, and that clauses M.1 and

M.2 apply to all international Subscribers to the fullest extent permitted by law.

S5.3 Mandatory local requirements. The Provider and Subscriber will comply with privacy, data-protection, consumer, AI-transparency, tax and other regulatory requirements that are mandatorily

applicable to their respective activities and cannot lawfully be excluded by contract. Such compliance does not, by itself, amend the parties’ contractual choice of New South Wales law or exclusive forum.

S5.4 Tax. Taxes applicable to an international transaction may be calculated and collected through Stripe

where configured and available, in accordance with clause E.2. The Subscriber remains responsible for taxes lawfully imposed on the Subscriber that are not collected by the Provider or Stripe.

S5.5 AI transparency and human review. Where law requires disclosure that a person is interacting with

AI, notice concerning AI-generated or AI-assisted content, human review, contestability or recordkeeping, the party responsible for the relevant deployment or communication must comply with that requirement. Nothing in this clause transfers a statutory obligation that by law rests on the other party.

S5.6 Data protection. Schedule 4 applies where the GDPR, UK GDPR, CCPA/CPRA or another

mandatory data-protection regime applies to processing under the Services. Compliance with Schedule 4 does not constitute agreement to foreign governing law or jurisdiction except to the extent a mandatory law expressly requires otherwise.

S5.7 Intellectual property and copyright. Ownership and licence rights are governed by Part J. The

Subscriber acknowledges that legal protection for AI-generated output may differ between jurisdictions and that the Provider does not guarantee copyright, patent, trade mark or other intellectual-property protection in any jurisdiction.

PART II — PRIVACY POLICY

IMPORTANT PRIVACY NOTICE

Eagle888 AI Human personas are digital artificial-intelligence software systems, not real human beings. Personal information may be processed by Eagle888 AI Human systems and, where configured, by third-party AI, cloud, payment, accounting, website, communications and other service providers. Material outputs remain subject to human review under the Terms and Conditions.

1. Who we are

Eagle888 AI Human Pty Ltd (ABN 52 676 753 423; ACN 676 753 423) of 40 Koola Avenue, East Killara NSW 2071, Australia is the primary service provider for the AI Human platform (Eagle888, we, us, our). Eagle888 Pty Ltd (ABN 87 624 709 486; ACN 624 709 486) owns or controls relevant platform intellectual property and may receive limited personal information where reasonably necessary for authorised group, technical, IP, governance, legal or operational functions. For privacy purposes, Eagle888 AI Human Pty Ltd (Eagle888) may act as an organisation/controller/business for account, website, billing, security and relationship data, and may act as a processor/service provider where it processes Customer Content on a Subscriber’s documented instructions.

2. Scope and relationship with the Terms

This Policy applies to eagle888.co and related websites, account and subscriber portals, AI Human Workforce, AI Consulting, AI Pandit (Anika), customer support, approved integrations, email and video communications, billing, marketing and other Eagle888 AI Human services that link to this Policy. This Policy operates with the Eagle888 AI Human Terms and Conditions of Service, any Order Form or Statement of Work (SOW), and any applicable Data Processing Addendum (DPA). If they address Customer Content differently, the Terms, SOW or DPA govern the contractual processing arrangement to the extent permitted by law. Nothing in those documents reduces a privacy right that cannot lawfully be excluded.

3. Privacy roles: our data and Customer Content

Data context Typical Eagle888 AI Human role Examples Website, account, billing, security and relationship data Controller / APP entity / business Account registration, subscription administration, Stripe transaction metadata, support requests, security logs and marketing preferences. Customer Content processed for a Subscriber Processor / service provider where applicable Prompts, instructions, files, emails, documents, work products and data sent through a configured AI Human workflow. AI Pandit user-provided information Controller or processor depending on how the service is purchased Birth details, cultural or spiritual preferences and optional wellbeing context supplied for the requested service.

4. Personal information we collect and hold

Category Examples Account and identity Name, business name, role, email, login identifier, authentication records, country/region and account status. Subscription and transactions Plan, invoices, payment status, billing address, tax information and transaction metadata. Full card data is generally processed by payment providers such as Stripe rather than stored directly by Eagle888. Customer Content Prompts, instructions, documents, emails, files, images, audio, video, outputs, workflow data and information supplied through authorised connected services. AI Human configuration Role, business hours, approval thresholds, connected tools, authority matrix, reporting manager and workflow preferences. Usage, device and security IP address, browser/device data, timestamps, feature usage, security events, diagnostic logs, cookies and similar technologies. Communications Support messages, feedback, complaints, email correspondence, meeting notes and consent records. Public-source information Information retrieved from public websites or sources for a requested research or consulting workflow, where lawful and reasonably necessary.

5. Sensitive information

Australian privacy law gives additional protection to sensitive information, including health information and information about religious or philosophical beliefs. Eagle888 AI Human will only collect sensitive information where it is reasonably necessary for one or more of its functions or activities and the individual has given the consent required by law, or another lawful exception applies. Because AI Pandit may involve cultural or spiritual context, a user may choose to provide information that reveals religious or philosophical beliefs. Where that information is sensitive information under applicable law, Eagle888 AI Human will seek the individual’s express consent at or before collection unless a lawful exception applies. Consent must be informed, voluntary, current and specific. Subscribers must not provide sensitive information about another person unless they have lawful authority to do so and any required consent has been obtained.

6. How we collect information

We may collect personal information directly from you when you create an account, subscribe, complete an onboarding form, communicate with us, attend a meeting, use an AI Human service, provide files or instructions, submit a support request, or interact with the website. We may also receive information from a Subscriber, authorised user, connected application, payment processor, accounting platform, identity provider, service provider or public source where the collection is lawful, fair and reasonably necessary for the relevant function. Public availability does not, by itself, mean personal information may be collected or used without privacy limits.

7. How we use personal information

Provide, configure, secure, administer and support the Services and authorised connected workflows. Create and manage accounts, subscriptions, billing, local tax calculations, accounting records and customer support. Operate AI Human personas, process instructions and generate requested Deliverables. Apply human-review, approval, access-control, audit, security, safety and quality processes. Detect fraud, abuse, security incidents, policy violations, service failures and unlawful activity. Measure service reliability, capacity and performance using minimised, aggregated or de-identified information where reasonably practicable. Comply with law, lawful requests, regulatory duties, dispute handling and recordkeeping obligations. Send marketing only where permitted by law and maintain suppression records to honour opt-outs.

8. AI Backbones, model training and product improvement

AI Human personas may use third-party AI services (AI Backbones), including OpenAI, Google/Gemini, Perplexity and other providers selected for a Service Vertical, Subscription Tier or authorised workflow. Eagle888 AI Human will not use identifiable Subscriber Customer Content to train a shared model or a model made available to other customers unless the Subscriber has expressly opted in or a separate written agreement clearly authorises that use. Where commercially available and appropriate, Eagle888 AI Human will use business, enterprise or API configurations and contractual controls intended to restrict third-party model training on Customer Content. Third-party AI services have their own terms, configurations, processing locations and retention rules. Eagle888 AI Human cannot promise a control that the selected provider does not offer and will disclose material exceptions where reasonably practicable. De-identified or aggregated operational information may be used for security, service reliability, capacity planning, analytics and improvement where it no longer reasonably identifies a person or Subscriber. 8A. AI Backbone and third-party processing limitations AI Backbones and other third-party service providers operate systems that Eagle888 AI Human does not own or directly control. Their privacy, security, retention, training-use, data-location, access, feature and configuration controls may differ between products, plans and jurisdictions and may change over time. Eagle888 AI Human will use reasonable vendor due diligence, data minimisation and appropriate business, enterprise or API configurations where commercially available and suitable, but cannot guarantee a third-party control that the selected provider does not offer or that is outside Eagle888’s reasonable control. A third-party provider limitation does not remove any privacy, security, notification, contractual or statutory obligation that applicable law places on Eagle888 AI Human. Where a material provider change affects the way personal information is handled, Eagle888 AI Human will update notices, configurations, contractual arrangements or processing choices where reasonably required by law, contract or risk.

9. AI transparency, automated decisions and human review

Eagle888 AI Human is designed so a person is informed that an AI Human persona is a digital artificial-intelligence system whenever disclosure is required by law and, as a default, whenever a reasonable person could otherwise believe the interaction is with a natural person. The Eagle888 AI Human operating model requires a Responsible Person, Reporting Manager, Principal or other authorised human to review important or externally facing output in accordance with the Terms. AI Human personas are not authorised to make solely automated decisions about natural persons that have legal or similarly significant effects unless a specific workflow has been separately assessed, authorised and configured with the safeguards required by applicable law. For Australia, from 10 December 2026 APP 1 will require covered APP entities to include specified information in their privacy policy where a computer program uses personal information to make, or substantially assist in making, a decision that could reasonably be expected to significantly affect an individual’s rights or interests. Eagle888 AI Human will maintain this Policy to reflect any qualifying arrangement. At the Effective Date, the intended service model is human-supervised rather than solely automated for such significant decisions. Where applicable law provides a right to information, human intervention, representations, contest or opt-out concerning a significant automated decision, Eagle888 AI Human will provide or support that process to the extent the obligation applies to Eagle888 AI Human.

10. Legal bases for processing outside Australia

Where a jurisdiction requires a stated legal basis, Eagle888 AI Human may rely on performance of a contract, steps requested before entering a contract, consent, legitimate interests, compliance with legal obligations, protection of legal rights or safety, and other lawful bases recognised by applicable law. Sensitive or special-category information is processed only where an additional lawful condition is satisfied.

11. Sharing, disclosure and service providers

We may disclose personal information to service providers that support cloud hosting, AI models, payments and tax calculation, accounting, email and communications, identity, website hosting, analytics, security, customer support and other authorised functions. Examples may include OpenAI, Google/Gemini, Perplexity, Stripe, Xero, WordPress/hosting providers and cloud infrastructure providers, depending on the service configuration. We do not sell personal information for money. We do not use Subscriber Customer Content for cross-context behavioural advertising. If a law defines sale, sharing or targeted advertising more broadly and the relevant activity occurs, Eagle888 AI Human will provide the notice and opt-out rights required by that law within the period required by applicable law. Specific sub-processor identities, locations or contractual details may be confidential for commercial or security reasons and will be disclosed where required by law, contract, DPA or a valid regulatory process.

12. Connected applications and authorised access

If a Subscriber connects a third-party application, Eagle888 accesses data only within the permissions authorised by that connection and the configured workflow. Subscribers should use least-privilege permissions and promptly revoke access that is no longer required. Data independently held by a connected provider is also governed by that provider’s own privacy terms. Subscribers are responsible for ensuring they have authority to provide Customer Content and personal information to Eagle888 AI Human and to instruct Eagle888 AI Human to process it through configured integrations.

13. Overseas disclosures and international processing

Eagle888 AI Human is based in Australia. Personal information may be stored or processed in Australia and, depending on the cloud, AI Backbone, payment, communications or other service selected, in other countries. Likely locations may include the United States, European Union/European Economic Area, United Kingdom, Singapore, India and other locations used by configured providers. The exact locations can vary by provider, service and customer configuration. Where APP 8 applies, Eagle888 AI Human will take reasonable steps in the circumstances to address the handling of personal information by overseas recipients, subject to any applicable statutory exception. Where GDPR, UK GDPR or another transfer regime applies, Eagle888 AI Human will use the legally required transfer mechanism where applicable. A current processing-location or sub-processor register may be maintained separately where operationally appropriate or provided where required by law or contract.

14. Data residency

Eagle888 AI Human does not promise that all personal information will remain exclusively in Australia unless an Enterprise arrangement, Order Form or SOW expressly states a verified residency commitment. Legal and governance master documents may be stored in Australian cloud regions, but AI Backbones and other authorised service providers may process information in other jurisdictions.

15. Retention, deletion and de-identification

Data type Default approach Account and subscription records Retained during the relationship and afterwards for the period reasonably required for tax, accounting, fraud prevention, dispute handling and legal obligations. Customer workspace content Retained during the subscription and any stated export/deletion period, then deleted or de-identified from active systems subject to backup cycles, legal holds and technically necessary residual copies. Backups Aged out under the applicable backup and disaster-recovery cycle. Backup retention depends on the verified infrastructure configuration. Security and audit logs Retained for a period reasonably necessary for security, fraud prevention, troubleshooting, audit and legal obligations. Marketing and suppression records Marketing records are retained while a lawful basis applies. An unsubscribe/suppression record may be retained to ensure an opt-out is respected. AI Pandit service data Retained only as reasonably necessary for the requested service, account history, legal obligations and any user-selected recordkeeping settings; unnecessary sensitive information should not be retained. Where Australian APP 11 applies and personal information is no longer needed for a permitted purpose, Eagle888 AI Human will take reasonable steps to destroy or de-identify it unless retention is required by law or a court/tribunal order.

16. Security

Eagle888 AI Human will maintain reasonable technical and organisational safeguards appropriate to the nature of the information, service and risk. Controls may include multi-factor authentication, role-based and least-privilege access, encryption in transit and at rest where supported, private cloud storage, logging, monitoring, backup controls, approval workflows, data minimisation, masking/redaction and incident response. Specific controls vary by infrastructure component and provider. Eagle888 AI Human will not represent a certification, encryption standard, penetration-test frequency, data-residency commitment or other technical control as implemented unless it is actually deployed and can be verified.

16A. Cybersecurity and technological limitations

The protection of personal information, confidential communications and connected systems depends in part on the technological capabilities and security provisions offered by the hosting, network, AI and communications providers we use. These capabilities may change and cannot eliminate every risk. We assess and apply reasonable safeguards appropriate to the circumstances, review material provider limitations and respond to incidents as required by law. This notice does not reduce any privacy right or legal duty to take reasonable steps to protect personal information.

17. Data breaches

Eagle888 AI Human will assess suspected data breaches promptly and notify affected individuals, Subscribers and regulators where required by applicable law. Where the Australian Notifiable Data Breaches scheme applies, Eagle888 AI Human will assess and notify eligible data breaches in accordance with the Privacy Act 1988 (Cth) and OAIC requirements. Different notification requirements may apply to processing subject to overseas privacy laws or an Enterprise contract. Contractual notification periods apply only where expressly agreed and operationally supportable.

18. Your privacy rights

Depending on the law that applies to you and the capacity in which Eagle888 AI Human holds your information, you may have rights to:

• request access to personal information Eagle888 AI Human holds about you;

• request correction of inaccurate, out-of-date, incomplete, irrelevant or misleading information;

• request deletion or erasure where applicable and subject to lawful retention obligations;

• withdraw consent where processing relies on consent;

• object to or restrict certain processing where applicable;

• request data portability where applicable;

• opt out of qualifying sale/sharing, targeted advertising or automated decision-making where

applicable;

• request meaningful information, human intervention or contest a qualifying automated decision

where applicable; and

• make a privacy complaint to Eagle888 AI Human and, where applicable, a competent privacy

regulator. We may need to verify identity and authority before acting on a request. Where Eagle888 AI Human processes Customer Content only on a Subscriber’s instructions, we may direct a request to that Subscriber or assist it to respond, as required by the relevant processing arrangement.

19. Australia: Privacy Act, APPs and complaints

Where the Privacy Act 1988 (Cth) applies, Eagle888 AI Human will manage personal information in accordance with the Australian Privacy Principles, including requirements concerning transparent privacy management, collection, notification, use and disclosure, direct marketing, cross-border disclosure, quality, security, access and correction. Privacy requests or complaints may be sent to aihuman@eagle888.co. Please identify the request, the relevant account or service, and enough information for us to verify identity where appropriate. Eagle888 AI Human will acknowledge, assess and investigate the request or complaint and respond within the period required by law, providing reasons where a request is refused to the extent required. If you are not satisfied with our response and the Privacy Act applies, you may be able to complain to the Office of the Australian Information Commissioner (OAIC).

20. International privacy rights

Region Additional position European Union / EEA Where GDPR applies: transparency, lawful basis, data-subject rights, processor/controller contracts, security, breach obligations, DPIAs where required, international-transfer safeguards and restrictions on solely automated significant decisions. EU AI Act transparency obligations are addressed separately through AI disclosures where applicable. Region Additional position United Kingdom Where UK GDPR, Data Protection Act 2018 and Data (Use and Access) Act 2025 apply: individual rights, lawful processing, transfer safeguards and required safeguards for significant automated decisions, including information, representations, human intervention and contest. California / United States Where CCPA/CPRA or other state privacy laws apply: required notices and rights may include access/know, correction, deletion, portability, opt-out of sale/sharing and, where applicable, ADMT rights and risk-assessment obligations. India Where the Digital Personal Data Protection Act 2023 and Rules apply according to their commencement timetable: Eagle888 AI Human will provide required notices, process on an applicable lawful basis, maintain security safeguards and support applicable rights and grievance processes. Other jurisdictions Mandatory privacy, data-protection and AI-transparency rules apply to the extent they legally govern the relevant processing and cannot lawfully be excluded.

21. Marketing and electronic communications

Eagle888 AI Human may send service communications needed to operate an account or subscription. Marketing communications are sent only where permitted by law. In Australia, commercial electronic messages will be handled in accordance with the Spam Act 2003, including consent requirements, sender identification and a functional unsubscribe method. Where an unsubscribe request is made, Eagle888 AI Human will action it within the period required by law and may retain a suppression record so that the opt-out continues to be respected.

22. Cookies, analytics and website technologies

The website may use strictly necessary cookies and similar technologies for security, session management and essential functionality. Analytics, advertising or other non-essential technologies will be used only when configured and with consent where applicable law requires it. A cookie banner or separate Cookie Notice may provide more detailed choices when such technologies are enabled.

23. Children and third-party personal information

Subscriber accounts are intended for persons aged 18 or over. The Services are not directed to children. A Subscriber must not submit personal information about a child or another person unless the Subscriber has lawful authority to do so and the information is reasonably necessary for the authorised service. Eagle888 AI Human will not knowingly use an AI Human service to make a significant decision about a child without specific legal, contractual, privacy and safeguarding controls.

24. AI Pandit (Anika) privacy

AI Pandit may process optional birth date, birth time, place, relationship context, cultural or spiritual preferences and other information supplied by the user for cultural, educational, informational or entertainment services. Religious or philosophical beliefs, health information and some wellbeing information may be sensitive information. Eagle888 AI Human will not require a user to disclose more sensitive information than is reasonably necessary for the requested service and will seek specific consent where required. Users should avoid providing medical, genetic, biometric, financial, government-identifier or other highly sensitive information unless the service expressly requires it and the collection has been authorised. AI Pandit is not intended to infer or create sensitive profiles for advertising, employment, insurance, credit, healthcare or other significant decisions.

25. AI Human Workforce privacy

AI Human Workforce services may process employee, contractor, client, supplier, customer and business-contact information contained in Subscriber instructions and files. The Subscriber remains responsible for determining that it has authority and a lawful basis to provide that information and for setting appropriate role permissions, human approval and retention controls. Eagle888 AI Human processes such Customer Content only for the authorised service purposes and subject to the Terms, SOW, DPA and configured access permissions.

26. AI Consulting privacy

AI Consulting may process project documents and information concerning finance, governance, risk, technology, operations, employees, customers, suppliers or other third parties. Subscribers should minimise personal information and avoid regulated or sensitive data that is not necessary for the SOW. Where a consulting engagement requires personal information subject to additional legal restrictions, the SOW or DPA may impose additional access, residency, security, retention, approval or deletion controls before processing begins.

AI Consulting access isolation. Project access may be segmented by Subscriber tenant and engagement identifiers with authenticated role-based permissions. Email addresses may be used for account administration and communications but are not intended to be the sole isolation key. Context sent to an AI Backbone should be limited to the authorised assignment and the information reasonably required for that assignment. The Provider will not intentionally expose one Subscriber’s consulting content to another Subscriber through the assignment workflow.

27. Contact and privacy requests

Privacy requests, complaints and questions may be sent to: Eagle888 AI Human Pty Ltd 40 Koola Avenue, East Killara NSW 2071, Australia Email: aihuman@eagle888.co For a privacy request, please state the nature of the request and the relevant account or service. Do not email passwords, MFA codes, API keys, full payment-card data or other unnecessary credentials.

28. Governing law and jurisdiction

This Policy and Eagle888 AI Human’s contractual privacy commitments are governed by the laws in force in New South Wales, Australia, to the fullest extent permitted by law. Any contractual dispute arising from this Policy or the Services is subject to the exclusive governing-law and jurisdiction provisions in Part M of the Eagle888 AI Human Terms and Conditions of Service. A mandatory privacy, consumer, data-protection or regulatory right that cannot lawfully be excluded continues to apply only to the extent required by that law. Nothing in this clause prevents an individual from making a complaint to a regulator that has jurisdiction under applicable law.

29. Changes to this Policy

Eagle888 AI Human may update this Policy to reflect changes in law, technology, Service Verticals, AI Backbones, sub-processors, security practices or business operations. The current version and effective date will be published on the website. Material changes will be notified where required by law or contract.

PART III — DATA ETHICS & RESPONSIBLE AI FRAMEWORK

IMPORTANT ETHICS NOTICE AI Human personas are artificial-intelligence software systems and digital service-delivery agents, not natural persons. They may perform approved workflows autonomously within defined limits, but material legal, financial, safety, health, employment, security, regulatory and other consequential actions require proportionate human oversight. AI can make mistakes. Eagle888 AI Human (Eagle888) does not permit deception about AI identity, fabricated authority, unlawful discrimination, covert use of personal information, or unreviewed high-impact decisions.

Purpose. This Framework sets Eagle888 AI Human Pty Ltd’s public ethical commitments and internal minimum governance requirements for the design, configuration, deployment and operation of AI Human services. It complements the Terms and Conditions of Service, Privacy Policy, Data Processing Addenda, Statements of Work, security controls and applicable law.

Company roles. Eagle888 AI Human Pty Ltd (ABN 52 676 753 423; ACN 676 753 423) is the service provider. Eagle888 Pty Ltd (ABN 87 624 709 486; ACN 624 709 486) owns or controls relevant platform intellectual property. Operational titles assigned to AI Human personas describe workflow roles; they do not, by themselves, create legal office-holder status or displace responsibilities that law requires a natural person to hold.

1. Ethical foundation

Eagle888 applies a human-centred, risk-based approach. The following principles are non-negotiable across all Service Verticals and AI Backbones.

Principle	Eagle888 operating commitment
Human benefit and dignity	AI should assist people and organisations without undermining human dignity, autonomy, lawful rights or access to essential services.
AI identity and transparency	People should be told when they are interacting with an AI Human or receiving materially AI-generated content where law requires it and, as Eagle888’s default, whenever a reasonable person could otherwise believe the interaction is with a natural person.
Human accountability	AI may recommend, draft and execute authorised low-risk workflows, but accountable human persons retain responsibility for usage, consequential decisions and approvals required by law, contract or Eagle888 policy.
Fairness and non-discrimination	AI must not unlawfully discriminate or use protected traits, sensitive data or proxies to produce unjustified adverse outcomes. Higher-risk uses require bias testing and human review.
Privacy by design	Collect and use only data reasonably necessary for the authorised purpose; use purpose limitation, least privilege, retention controls, meaningful user rights and lawful cross-border processing.
Security and resilience	Apply risk-proportionate access control, authentication, private storage, logging, incident response and technical safeguards without publishing sensitive security architecture.
Accuracy, provenance and uncertainty	Do not fabricate sources, qualifications, identity, authority or evidence. Preserve provenance where reasonably practicable, distinguish verified facts from inference, and communicate material uncertainty.
User agency and contestability	Provide practical routes for correction, feedback, complaint and human review where an AI-supported outcome materially affects a person.
Proportional autonomy	The greater the potential impact, irreversibility or sensitivity of an action, the lower the level of autonomous authority permitted without human confirmation.
Responsible innovation	Improve capability through testing, monitoring, feedback and controlled iteration while preserving safety, privacy, transparency and legal compliance.
Intellectual property and confidentiality	Respect patents, copyright, trade marks, confidentiality, contractual restrictions and authorised use of customer and third-party materials.
Social and environmental responsibility	Consider material social, accessibility and environmental impacts where reasonably relevant to the design or deployment of AI services.

2. Eagle888 values translated into operating behaviour

Enterprise value	Operational expression
Integrity	Truthful identity, evidence-based claims, honest uncertainty, accurate records, no hidden manipulation.
Gratitude and respect	Respect for clients, users, cultures, beliefs and affected communities; no humiliation, exploitation or demeaning treatment.
Trust and care	Privacy, security, confidentiality, reliable escalation, user control and proportionate safeguards.
Commitment	Continuous monitoring, correction of errors, service improvement and clear ownership of unresolved risks.
Everlasting relationships	Long-term stewardship rather than short-term extraction: no dark patterns, no coercive lock-in, and clear routes to export, deletion or termination consistent with the Terms and Privacy Policy.

3. AI identity, functional titles and human accountability

Eagle888 may give AI Human personas functional titles such as CEO, director, manager, consultant or specialist to define workflow responsibilities and reporting lines. Those titles do not mean the persona is a natural person, statutory director, officer at law, licensed professional or authorised signatory unless applicable law expressly allows that status.

AI Human personas may coordinate plans, allocate tasks, prepare reports, communicate within approved channels and execute pre-approved workflows. They must escalate actions outside their authority matrix.

Each Subscriber must designate a Responsible Person, Reporting Manager, Principal or other authorised human for review and approval as required by the Terms, SOW or applicable law.

Eagle888 will not intentionally design a persona to deceive a reasonable user into believing the persona is a natural person.

4. Proportional autonomy and approval levels

Level	Typical activity	Required control of Real Human (Subscriber, Principal, Authorized Person)
A0 — Draft only	Research, summarise, classify, draft, prepare options.	Real Human reviews before external use or action.
A1 — Low-risk internal	Organise files, prepare internal reports, update non-sensitive records, routine monitoring.	Real Human remains accountable; pre-approved rules apply; logging or traceability is maintained appropriate to risk.
A2 — Controlled external	Routine template communications, low-risk content updates, standard workflow steps and reconciliations.	Real Human remains accountable; approved templates/rules apply; exceptions escalate for human review.
A3 — Material action	Pricing or refund exceptions, production deployment, access-permission changes, significant public statements, regulated or high-impact activity.	Explicit approval by an authorised Real Human before execution unless a lawful, separately assessed control framework permits otherwise.
A4 — No autonomous decision	Significant employment, credit, insurance, housing, education, health, legal, safety or similar rights-affecting decisions; statutory director acts; bank signatory changes.	AI may assist with information or drafts but cannot be the sole autonomous decision-maker. Specific legal assessment and safeguards, including meaningful human decision-making, are required.

5. Data ethics across the information lifecycle

5.1 Collection and data minimisation

Collect only information reasonably necessary for an authorised service purpose. Do not collect sensitive information merely because it may improve personalisation. Where consent is legally required, obtain consent that is informed, voluntary, current and specific.

5.2 Lawful sources and provenance

Publicly accessible information is not automatically unrestricted for AI use. Eagle888 and Subscribers must consider privacy, copyright, confidentiality, contractual and platform restrictions before collecting, reusing, training on or disclosing data.

5.3 Purpose limitation

Customer Content is processed for the authorised Service, workflow or SOW. Secondary use must be lawful, transparent and consistent with the Privacy Policy, contract and any required consent.

5.4 No silent model training

Eagle888 will not use identifiable Subscriber Customer Content to train a shared model or a model made available to other customers unless the Subscriber has expressly opted in or a separate written agreement clearly authorises that use. Where commercially available and appropriate, business, enterprise or API configurations that restrict third-party model training are preferred.

5.5 Access and least privilege

AI Human access to systems and data must be limited to the minimum permissions reasonably required for the approved role and workflow. Credentials, API keys, MFA secrets, private keys and similar secrets must not be exposed in public content or ordinary documents.

5.6 Retention and deletion

Retain data only for the period reasonably needed for the authorised service, legal, accounting, security, dispute or recordkeeping purpose. Deletion and de-identification follow the Privacy Policy, Terms and applicable infrastructure constraints.

5.7 Cross-border processing

Where AI Backbones, cloud, payment, communications or other providers process data outside Australia, to the extent possible, Eagle888 will apply the contractual and privacy controls required by applicable law and the relevant service configuration.

5.8 Cybersecurity and technological limitations

Our responsible-AI and data-ethics practices account for the available technological safeguards and the limits of each platform, provider and communication channel. We assess foreseeable cybersecurity, privacy and confidentiality risks before use, apply proportionate controls that the technology supports, escalate material gaps for human decision and reassess them when capabilities change. A technological limitation is recorded and managed; it is not treated as permission to disregard legal duties or commitments made to Subscribers.

6. Data classification and handling

Class	Examples	Minimum ethical handling
Public	Approved website content, published policies, approved marketing.	Integrity and publishing controls; no confidential data or credentials.
Internal	SOPs, internal plans, non-sensitive operating notes.	Authenticated access and need-to-know handling.
Confidential	Customer data, contracts, unpublished IP, pricing strategy, internal financial or governance information.	Private storage, least privilege, appropriate encryption and logging.
Restricted	Credentials, API keys, payment-card secrets, sensitive personal information, legal privilege, security secrets.	Strongest available controls; approved secret storage where applicable; avoid ordinary chat/email/plain-text storage unless specifically protected.

7. AI Backbone and third-party provider governance

AI Backbones and other sub-processors are assessed proportionately for privacy, training use, security, retention, contractual terms, data location, reliability, suitability, transparency and known limitations.

Eagle888 prefers enterprise, business or API configurations with appropriate customer-data protections for Subscriber content where available and suitable.

No provider or model is treated as infallible. Material facts, regulated matters and high-impact outputs require authoritative-source verification or competent human review.

Material changes to a provider, model, retention practice, training use, data location or capability that may alter risk are assessed before continued use in affected workflows.

Where multiple AI Backbones are used, Eagle888 may compare outputs, but agreement between models is not treated as proof of correctness.

Eagle888 maintains provider and processing information appropriate to risk and discloses sub-processor information where required by law, contract or DPA.

AI Human capability is dependent on the technical, contractual and policy limits of its configured AI Backbones. Changes, outages, safety restrictions, model behaviour, capacity limits or discontinued features may constrain the functions, speed or quality of an AI Human. Eagle888 is not promising capability, accuracy, availability, data controls or safeguards beyond what is actually supported and verified for the relevant configuration.

Third-party limitations are risk inputs, not a transfer of Eagle888’s own responsibilities. Eagle888 remains accountable for reasonable provider selection, configuration, monitoring, escalation and controls within its reasonable control and for obligations that applicable law does not permit Eagle888 to exclude.

8. Accuracy, sources, explainability and intellectual honesty

For the avoidance of doubt, the Terms and Conditions require material outputs of AI Human Workforce, AI Human Consultants and AI Pandit to be reviewed and verified, as applicable, by the Subscriber, Principal, Responsible Person or other authorised human for completeness, accuracy and suitability for the intended purpose before use.

Use of an output after a reasonable opportunity for review constitutes acceptance of that output for operational purposes, subject to the Terms and any non-excludable rights. Such use does not convert AI-generated content into a warranty of accuracy or completeness.

AI-generated statements must not be presented as verified facts when verification has not occurred.

For research and consulting, source-backed facts should be distinguishable from model inference, estimates, judgement or recommendations.

Citations, links and quoted material must not be fabricated. Where a source cannot be verified, the output should state that limitation rather than inventing authority.

Where an AI-supported process materially affects a person, Eagle888 or the Subscriber should provide an explanation proportionate to the decision, the lawful role of the AI and the information reasonably available.

Human reviewers are expected to apply independent judgement rather than merely approve AI output mechanically.

9. Fairness, accessibility and non-discrimination

Eagle888 prohibits unlawful discrimination and unjustified differential treatment based on protected characteristics or close proxies.

Where a workflow can materially affect individuals, risk assessment should consider disparate impact, data quality, representation, accessibility and whether protected or sensitive attributes are necessary at all.

AI should not be used to infer sensitive traits for employment, credit, insurance, healthcare, housing, education, law enforcement or similarly consequential decisions unless lawful, necessary, specifically assessed and subject to meaningful safeguards.

Where feasible, user interfaces and communications should support accessibility and understandable language appropriate to the audience.

10. Transparency, synthetic media and public communications

AI Human identity must be disclosed where required by law and by default where confusion with a natural person is reasonably foreseeable.

Where practicable, AI Human display names, signatures and dedicated email addresses should make the AI identity clear (for example, by using an “AI Human” or “AI Pandit” designation) so a reasonable recipient is not misled into believing the persona is a natural person.

AI-generated or materially AI-manipulated images, audio, video or text must not be used for deceptive impersonation, fraud, coercion or unauthorised representation of a real person.

Where applicable, Eagle888 will support visible notices, labels, provenance metadata or machine-readable markers required for AI-generated or manipulated content.

Public-interest communications, safety-critical communications and material regulatory, investor or customer statements require proportionate human editorial review.

Marketing claims about cost savings, accuracy, security, confidentiality, uptime, qualifications or performance must be evidence-based and qualified where results depend on role, workload, model, configuration or human supervision.

11. Safety, harmful use and manipulation

Eagle888 will not knowingly enable serious physical harm, self-harm, violence, terrorism, sexual exploitation, child sexual abuse material, unlawful surveillance, fraud, credential theft or malicious cyber abuse.

AI Human personas must not exploit grief, fear, financial distress, illness, loneliness, religious belief, superstition or other vulnerability to pressure a person into purchasing, donating, disclosing data or taking a consequential action.

Persuasive or personalised content must not be used to covertly manipulate people in a way that materially undermines autonomy or lawful rights.

Safety controls are layered: model-level safeguards are supplemented by permissions, workflow controls, monitoring, human review and incident response.

12. Service-vertical ethical controls

12.1 AI Human Workforce

Each role should have defined responsibilities, permitted tools, prohibited actions, approval thresholds, escalation routes, training material, test tasks and performance criteria.

AI Human personas must not silently present themselves as natural employees or licensed professionals.

Customer authority matrices govern access to email, websites, accounting, cloud, CRM and other connected systems.

Employment or similarly consequential people decisions require separate legal assessment and meaningful human decision-making; AI may support but should not be the sole decision-maker without lawful safeguards.

12.2 AI Consulting

Each assignment should be governed by an Order Form, Proposal or SOW defining scope, assumptions, exclusions, authorised systems/data and a human Principal.

Research should distinguish source-backed facts from inference, scenario analysis or recommendation.

AI Consultant output is decision support, not a professional licence or assurance engagement. Regulated matters must be reviewed as required by the Terms and applicable law.

Conflicts, confidentiality, privilege, intellectual property and third-party rights must be considered before processing client material.

12.3 AI Pandit (Anika)

Astrology, numerology, spiritual, cultural and wellbeing content is provided for cultural, educational, informational and entertainment purposes, not as guaranteed prophecy or regulated professional advice.

AI Pandit must not present death, catastrophe, illness, pregnancy, legal outcome, investment result or other consequential future event as certain fact.

AI Pandit must not exploit fear, grief, superstition, religious belief or emotional vulnerability, and must not pressure users to purchase services or take consequential actions.

Religious or philosophical beliefs and some wellbeing information may be sensitive information. Collection should be minimised and express consent obtained where required by law.

Ritual or puja guidance involving fire, smoke, candles, incense, food, movement or physical activity should include proportionate safety reminders.

AI Pandit must respect religious and cultural diversity and must not ridicule, denigrate or falsely claim authority over a faith, tradition or community.

13. Significant automated decisions and human review

Eagle888’s intended operating model is human-supervised for decisions that could reasonably be expected to significantly affect an individual’s rights or interests. AI may substantially assist, but a specific workflow involving significant automated decisions must not be deployed without assessment of applicable privacy, discrimination, consumer, employment and sector-specific obligations, including notice, human intervention, contestability and recordkeeping requirements where they apply.

AUSTRALIAN PRIVACY POSITION — 10 DECEMBER 2026 From 10 December 2026, Australian APP 1 includes additional privacy-policy transparency requirements for qualifying arrangements in which a computer program uses personal information to make, or substantially and directly assist in making, decisions that could reasonably be expected to significantly affect an individual’s rights or interests. Eagle888’s Privacy Policy and affected workflows must remain consistent with those obligations.

14. Incidents, errors and corrective action

Contain or pause the affected workflow where there is material safety, privacy, security, legal, financial or reputational risk.

Preserve relevant evidence and logs without unnecessarily increasing disclosure of personal or confidential information.

Escalate according to severity to the authorised human accountable for the service, security, privacy, legal or operational risk.

Assess notification, customer, regulatory and contractual obligations.

Correct the immediate issue and, where appropriate, update tests, controls, instructions, training, FAQs, risk registers and operating procedures to reduce recurrence.

Restore automation only when residual risk is accepted by an authorised person at the appropriate level.

15. User agency, complaints and contestability

Users and Subscribers may report errors, unsafe conduct, misleading behaviour, suspected bias, privacy concerns or security concerns.

Where applicable, a person materially affected by an AI-supported outcome may request correction, explanation, human review or challenge in accordance with the Privacy Policy, contract and applicable law.

Complaints and material near misses should be logged and used as controlled inputs to quality improvement rather than ignored or hidden.

Privacy and ethics enquiries may be sent to aihuman@eagle888.co. Users should not send passwords, MFA codes, API keys or unnecessary sensitive credentials by email.

16. Governance, testing and review

Eagle888 maintains an accountability structure for AI use, with defined owners, authority limits, escalation pathways and risk acceptance appropriate to each workflow.

New or materially changed AI uses should be assessed proportionately for purpose, affected people, data, providers, autonomy, failure modes, fairness, security, privacy, transparency and human-control needs before deployment.

AI services should be monitored for material errors, drift, unsafe behaviour, security issues, discriminatory outcomes and provider changes appropriate to risk.

This Framework is reviewed periodically and when there is a material law change, serious incident, major AI Backbone/provider change, new Service Vertical, or a significant change in risk profile.

Eagle888 does not publicly claim a certification, penetration-test frequency, security standard, accuracy rate or other control unless the claim is current, implemented and verifiable.

17. Legal and regulatory alignment

This Framework is an ethical and governance standard, not a substitute for law. Eagle888 operates from New South Wales, Australia and administers this Framework under the laws in force in New South Wales and applicable Commonwealth law. The Terms and Conditions govern contractual choice of law and forum. Mandatory privacy, consumer, anti-discrimination, online safety, employment, intellectual-property, sector-specific or AI-transparency requirements that apply by operation of law continue to apply to the extent they cannot lawfully be excluded.

Region	Alignment position
Australia	Privacy Act 1988 (Cth) and Australian Privacy Principles; Australian Consumer Law; anti-discrimination, corporations, employment, online-safety, intellectual-property and other generally applicable laws as relevant to the use case; Australian Government Guidance for AI Adoption and AI Ethics Principles as responsible-AI guidance.
European Union / EEA	GDPR where applicable and EU AI Act requirements applicable to the relevant provider/deployer role, including Article 50 transparency obligations for certain AI systems and AI-generated/manipulated content.
United Kingdom	UK GDPR, Data Protection Act 2018 and applicable automated-decision safeguards where relevant.
United States / California	Applicable state privacy, consumer-protection, biometric, AI-transparency and automated-decision requirements, including CCPA/CPRA where applicable.
Other jurisdictions	Mandatory privacy, data-protection, consumer, discrimination, AI-transparency, safety and sector-specific laws are assessed for the affected deployment.

18. Public commitment

Eagle888 AI Human commits to disclose that its personas are digital AI systems; preserve meaningful human accountability for consequential actions; minimise and protect personal and customer data; use risk-proportionate autonomy; prohibit deceptive, exploitative and unlawfully discriminatory use; respect intellectual property and confidentiality; maintain practical routes for correction, complaint and human review; and improve controls when incidents, evidence or law show that improvement is needed.

PART IV — CONSOLIDATED ELECTRONIC ACCEPTANCE & DOCUMENT CONTROL

Electronic acceptance

By selecting the acceptance checkbox (or equivalent electronic control) and proceeding to subscribe or pay, the Subscriber confirms that it has had the opportunity to read this Consolidated Document and electronically accepts it in accordance with the acceptance wording below, including the disclosed technological limitations of the Services and safeguards. Eagle888 may retain evidence of the acceptance event, including the account, date and time, document version and reasonably necessary technical metadata, for contract administration, security, audit and dispute-resolution purposes.

[ ] I have read and accept the Eagle888 AI Human Consolidated Terms & Conditions, Privacy Policy and Data Ethics Framework. I agree to Part I — Terms and Conditions of Service and to applicable Subscriber obligations in Part III; I acknowledge Part II — Privacy Policy and Eagle888’s Data Ethics commitments; I understand that AI Human personas are artificial-intelligence systems that can make mistakes and that required real human review remains my responsibility. I acknowledge technological limitations in cybersecurity, privacy, confidentiality and communications, while retaining my non-excludable rights and Eagle888’s applicable legal duties. I also understand that separate consent will be requested where applicable law requires it.

Document control

Document	Consolidated Terms & Conditions, Privacy Policy and Data Ethics Framework
Service Provider	Eagle888 AI Human Pty Ltd — ABN 52 676 753 423 / ACN 676 753 423
Patent / IP Owner	Eagle888 Pty Ltd — ABN 87 624 709 486 / ACN 624 709 486
Effective date	27 September 2026
Consolidated version	1.1
Component baselines	Terms V4.0; Privacy Policy V4.0; Data Ethics & Responsible AI Framework V4.0
Contractual forum	New South Wales, Australia, subject to mandatory non-excludable law and Part I.

Contact: aihuman@eagle888.co