Skip to content
GLOBALANCE INTELLIGENCEThe economy ahead.
This Week Index 28/100 Sign in
Money & Investments › GI 100 Index
The Great Value ShiftInvestment ThemesGI 100 IndexGI PortfolioStart Future Check
AI FOOTPRINTsolidASSESSMENT 2026GLOBALANCEINTELLIGENCE
Globalance AI Footprint

Assessment as of November 2026 · next assessment 2027.

Result of the assessment by the Globalance research team under the published methodology. Not a certification or quality label.

How we assess ›

How this rating is builtVideo · 0:52
GI 100 Index · Profile

Airbnb Inc

USAAdoptersAssignment FullData quality High

Key messagesolid AI Footprint

Airbnb uses artificial intelligence widely, for example in customer service and in screening risky bookings, and informs users about these uses and about the option to object to the use of their data for training. Clear published rules and responsibilities for handling artificial intelligence are lacking so far, and no independent testing of the systems was found.

Opportunities

  • Safetysolid

    Concrete procedures for protection against misuse and identity abuse are documented: automated detection of fake accounts, listings and reviews, and reservation screening with possible manual review and an appeal route (S22, S33), plus a coordinated vulnerability disclosure procedure via HackerOne …

    Concrete procedures for protection against misuse and identity abuse are documented: automated detection of fake accounts, listings and reviews, and reservation screening with possible manual review and an appeal route (S22, S33), plus a coordinated vulnerability disclosure procedure via HackerOne (S36) and board oversight of the cyber risk programme (S08); Item 1C in the primary text confirms an incident response plan, external assessment providers, training and third-party risk management, but names neither a framework nor AI (S47). For the AI customer service agent there are only the company's own descriptions of evaluation and protection mechanisms (S16, S17), no independent testing. The counter-evidence is minor: a DPC reprimand for a disproportionate ID requirement with documented remediation (S01), a case of AI-manipulated damage photos that Airbnb's review process initially accepted (S03), and unconfirmed reports of erroneous actions by the support agent (S39). The rating “solid” rests on the company's own documentation (grade B) and is flagged for review.

  • Transparencysolid

    Airbnb Seals AI answers in the Help Center, points out possible errors and discloses the areas in which AI is used and the use of personal data for model training, including an opt-out (S41, S12).

    Airbnb Seals AI answers in the Help Center, points out possible errors and discloses the areas in which AI is used and the use of personal data for model training, including an opt-out (S41, S12). The DSA transparency report describes the automated means and complaint figures but gives no error rates (S22); the help page on reservation screening states signal categories and consequences (S33). Explainability is limited: the privacy supplement denies automated decisions with significant effects, although the screening systems can block bookings and remove accounts (S34, S33), and bans are sometimes imposed without reasons (S29); which third-party models are used is known only from statements by the CEO (S24, S26).

Risks

  • Responsibilitylimited

    The documented board oversight (Audit, Risk and Compliance Committee) covers cybersecurity, privacy and IT risks; a specific responsibility for AI is not apparent in the part of the proxy statement that was read (S08, S15).

    The documented board oversight (Audit, Risk and Compliance Committee) covers cybersecurity, privacy and IT risks; a specific responsibility for AI is not apparent in the part of the proxy statement that was read (S08, S15). A published AI policy, risk classification or approval process for in-house and purchased models is missing; there is only the declaration of intent “committed to the responsible development and use of AI” (S12) and indications from a job advertisement of privacy reviews for AI (S38). Complaint channels are documented concretely in the DSA report and the help pages (S22, S33), but they relate to platform moderation in general and not to AI governance; hence “limited” at the level of principles.

  • Risk area

    Automated reservation screening that can block bookings and remove accounts

  • Risk area

    AI customer service with authority over refunds and rebookings

  • Risk area

    Identity verification and fraud detection using ID data

  • No serious controversy recorded.

What lies behind the assessment

Research of 3 November 2026
41searches
46pages and documents read
11primary documents
6controversies checked
5open points, named openly
The full research: pillars, evidence, controversies and open points
AI Footprint · Adopter · As of 3 November 2026
Airbnb IncOverall rating: solid

Airbnb uses AI in customer service and in screening risky bookings and informs users about it. A published AI policy, board-level responsibility for AI and independent testing are missing.

Responsibility

40 %
limitedsolidstrong

The board oversees cyber and privacy risks, but responsibility for AI is missing.

  • +A board committee oversees cyber and privacy risks Source ↗
  • +Complaints channel with human review according to the DSA report Source ↗
  • –No AI responsibility documented on the board or in a committee Source ↗
  • –No published AI policy, only a declaration of intent Source ↗
  • –No approval process documented for some 13 purchased models Source ↗
All sources for this pillar (14)

Safety

40 %
limitedsolidstrong

Concrete procedures against misuse are documented, but independent testing of the AI agent is missing.

  • +Automated detection of fake accounts, listings and reviews Source ↗
  • +Reservation screening with manual review and an appeal route Source ↗
  • +Vulnerability reporting via HackerOne Source ↗
  • –Only self-descriptions of AI customer service, no independent testing Source ↗
  • –AI-manipulated damage photos were initially accepted Source ↗
All sources for this pillar (16)
What is missing for “strong”

These steps are in the company’s own hands.

1Responsibility
Set AI rules and responsibility

Publish an AI policy with risk classification and an approval process for purchased models and assign AI oversight to a board committee.

2Safety
Have the AI agent independently tested

Have the AI customer service agent and reservation screening tested independently and publish error rates.

3Transparency
Explain automated bans

Give reasons for automatically triggered bans and disclose which foundation models are in use.

Context: 101 companies assessed

Distribution of overall ratings, no ranking
strong
none yet
solid
← Airbnb Inc is one of 35
limited

No company reaches “strong” today. Disclosure on how companies handle AI is still young across the market.

Controversies
6 6 minor

No deduction for controversies.

Largest risk area
Automated bans without clear explanation

Reservation screening can block bookings and remove accounts, while the privacy supplement denies making automated decisions with significant effects.

How thoroughly assessed
40 Evidence

11 primary documents read, 41 searches.

Read the full reasoning and all sources
Airbnb Inc · Full reasoning

Broad use of AI with notices for users, but without dedicated AI rules or responsibility

Overall rating “solid”. No deduction for controversies. The superscript numbers link directly to the source.

1 · Responsibility

limitedWeight 40%

Airbnb has committee oversight of cyber, privacy and IT risks, but no published AI policy and no approval process for purchased models.

What is documented
  • The Audit, Risk and Compliance Committee oversees cybersecurity, privacy and IT risks1
  • The Chief Privacy Officer, Chief Security Officer and Director Information Security report regularly to this committee2
  • The 10-K contains a dedicated risk factor on AI and names the EU AI Act as a regulatory risk7 8
  • According to the DSA report, 2,338 of 5,668 complaints against bans or restrictions were decided in favour of those affected12
  • Reservation screening decisions can be appealed via the guest inbox13
The case against
Missing AI governance
  • No AI-specific committee responsibility is apparent in the part of the proxy statement that was read1
  • There is no published AI policy, only the declaration of intent “committed to the responsible development and use of AI”11
  • No approval process is documented for the roughly 13 purchased models from external providers10
  • Privacy reviews for AI are apparent only from a job advertisement9
No AI accountability
  • No AI reports, AI metrics, AI training or external AI audits were found
  • The 2023 sustainability report does not mention AI2
Sources for Responsibility (14)
  1. Airbnb, Inc. (SEC EDGAR): Definitive Proxy Statement (DEF 14A) 2026 · 24 April 2026 · primary source
  2. Airbnb, Inc.: Sustainability and Community Update 2023 · 16 November 2023 · secondary source
  3. Airbnb, Inc. (SEC EDGAR): Form 10-K for fiscal year 2025, Item 1C Cybersecurity (XBRL page R28 “Cybersecurity Risk Management and Strategy Disclosure”) · 12 February 2026 · primary source
  4. Skift: Airbnb Hires Former Meta AI Chief Ahmad Al-Dahle as CTO · 14 January 2026 · secondary source
  5. The Motley Fool: Airbnb (ABNB) Q4 2025 Earnings Call Transcript · 12 February 2026 · secondary source
  6. Airbnb, Inc. (SEC EDGAR): Form 10-K for fiscal year 2025 · 12 February 2026 · secondary source
  7. SEC EDGAR Full-Text Search: Search for “artificial intelligence and machine learning gives rise” (CIK 0001559720) · 3 November 2026 · secondary source
  8. Bloomberg Law: From Meta to Airbnb, Companies Flag Risks Dealing With EU AI Act · 3 March 2025 · secondary source
  9. Airbnb, Inc. (Careers): Senior Privacy Manager, AI/ML governance (job advertisement) · 3 November 2026 · secondary source
  10. South China Morning Post: Airbnb picks Alibaba's Qwen over ChatGPT in a win for Chinese open-source AI · 22 October 2025 · secondary source
  11. Airbnb, Inc. (Help Center): AI-powered features on Airbnb (Help article 4097) · 3 November 2026 · primary source
  12. Airbnb Ireland UC: Transparency Report (DSA) for 2024-02-17 to 2024-12-31 · 14 February 2025 · primary source
  13. Airbnb, Inc. (Help Center): How Airbnb screens for potential disruptive gathering risk (Help article 3280) · 3 November 2026 · primary source
  14. Airbnb, Inc.: Airbnb 2024 Project Lighthouse Update · 12 December 2024 · primary source

2 · Safety

solidWeight 40%

Airbnb documents procedures against fake accounts, listings and risky bookings, but relies only on its own descriptions for its AI customer service.

What is documented
  • The DSA report describes automated means against fake accounts, bots, account takeovers, fake listings and fake reviews, with human review10
  • Reservation screening can trigger a manual review by staff; those affected can appeal11
  • Item 1C of the 10-K confirms in the primary text an incident response plan, external assessment providers, training and third-party risk management5 6
  • Vulnerabilities can be reported via HackerOne with coordinated disclosure7
  • Personal data is de-identified or pseudonymised before AI training8
  • Engineering posts describe evaluations with expert Seals, samples of live traffic and protection against prompt injection1 2
The case against
No independent testing
  • There are no independent tests or red-teaming reports on the AI customer service agent1 2
  • Error rates of reservation screening and other automated means are not published10
  • No security certifications were found
  • Item 1C does not mention AI and names no security framework5
Incidents and proceedings
  • In 2024 the Irish Data Protection Commission found infringements over a disproportionate ID requirement; remediation is documented9
  • Reports cite a host's AI-manipulated damage photos that Airbnb initially accepted; it refunded only after a media enquiry15
  • Two unconfirmed user reports describe a cancellation without confirmation and a triple refund by the AI support4
Sources for Safety (16)
  1. Airbnb Engineering & Data Science: Eval-driven development: Lessons from evaluating GenAI at scale · 3 November 2026 · secondary source
  2. QCon San Francisco: How Airbnb Guardrailed Its AI Customer Support Agent (conference talk abstract) · 16 November 2026 · secondary source
  3. Airbnb, Inc. (Help Center): AI answers in the Airbnb Help Center (Help article 4178) · 3 November 2026 · primary source
  4. Vibe Graveyard: Airbnb's AI support reportedly canceled a reservation and multiplied a host's refunds · 19 July 2026 · secondary source
  5. Airbnb, Inc. (SEC EDGAR): Form 10-K for fiscal year 2025, Item 1C Cybersecurity (XBRL page R28 “Cybersecurity Risk Management and Strategy Disclosure”) · 12 February 2026 · primary source
  6. Airbnb, Inc. (SEC EDGAR): Definitive Proxy Statement (DEF 14A) 2026 · 24 April 2026 · primary source
  7. Airbnb, Inc.: Security / Responsible Disclosure · 3 November 2026 · primary source
  8. Airbnb, Inc. (Help Center): AI-powered features on Airbnb (Help article 4097) · 3 November 2026 · primary source
  9. Data Protection Commission (Ireland): Inquiry into Airbnb Ireland UC – Decision 31 January 2024 (Reprimand) · 1 April 2024 · primary source
  10. Airbnb Ireland UC: Transparency Report (DSA) for 2024-02-17 to 2024-12-31 · 14 February 2025 · primary source
  11. Airbnb, Inc. (Help Center): How Airbnb screens for potential disruptive gathering risk (Help article 3280) · 3 November 2026 · primary source
  12. Airbnb, Inc. (Newsroom): Anti-party defenses for New Year's Eve rolled out around the world · 12 December 2024 · secondary source
  13. PYMNTS: Airbnb Deploys AI to Prevent Disruptive Parties at Rented Homes · 12 December 2025 · secondary source
  14. The Motley Fool: Airbnb (ABNB) Q4 2025 Earnings Call Transcript · 12 February 2026 · secondary source
  15. AI Incident Database: Report 6105 (Incident 1161): Airbnb host allegedly used AI-manipulated photos for a damage claim · 2 August 2025 · secondary source
  16. European Data Protection Board (EDPB) / DPC Ireland: Decision No 1364 – Amicable resolution, Airbnb Ireland UC · 1 January 2025 · primary source

3 · Transparency

solidWeight 20%

Airbnb Seals AI answers and discloses the use of data for training along with an opt-out, but explains automated bans only to a limited extent.

What is documented
  • AI-generated answers in the Help Center are labelled, cite sources and point out possible errors; switching to staff is possible at any time1
  • Airbnb lists the areas where AI is used and describes the use of personal data for model development with an opt-out2
  • The DSA report describes the automated means along the user journey3
  • The help page on reservation screening states signal categories and possible consequences5 6
The case against
Explainability of automated decisions
  • The privacy supplement denies automated decisions with significant effects, although screening can block bookings and remove accounts7 5
  • Reports describe a lifetime ban without reasons that was lifted only after intervention8
  • The DSA report gives no error rates for the automated means3
Model information
  • Which third-party models are used is known only from media reports on statements by the CEO9 10
  • No system description of the AI customer service agent was found
Sources for Transparency (10)
  1. Airbnb, Inc. (Help Center): AI answers in the Airbnb Help Center (Help article 4178) · 3 November 2026 · primary source
  2. Airbnb, Inc. (Help Center): AI-powered features on Airbnb (Help article 4097) · 3 November 2026 · primary source
  3. Airbnb Ireland UC: Transparency Report (DSA) for 2024-02-17 to 2024-12-31 · 14 February 2025 · primary source
  4. Rental Scale-Up: What Airbnb's New Privacy Terms Reveal About Its AI Plans for Summer 2026 · 1 April 2026 · secondary source
  5. Airbnb, Inc. (Help Center): How Airbnb screens for potential disruptive gathering risk (Help article 3280) · 3 November 2026 · primary source
  6. Airbnb, Inc. (Newsroom): Anti-party defenses for New Year's Eve rolled out around the world · 12 December 2024 · secondary source
  7. Airbnb, Inc. (Help Center): Privacy supplement (Help article 2863), last updated 2026-01-01 · 1 January 2026 · secondary source
  8. Elliott Report (Christopher Elliott): Help! Airbnb banned me for life but won't say why · 12 November 2025 · secondary source
  9. South China Morning Post: Airbnb picks Alibaba's Qwen over ChatGPT in a win for Chinese open-source AI · 22 October 2025 · secondary source
  10. The Motley Fool: Airbnb (ABNB) Q4 2025 Earnings Call Transcript · 12 February 2026 · secondary source

4 · Why the harm rule does not apply

The harm rule does not apply: within the 36-month window there is no judgment, regulatory decision or fine concerning harm caused by Airbnb's own AI. The AI-related cases are weakly documented individual cases, not a pattern.

5 · Controversies (6)

6 · What remains open

  • The full text of the AI risk factor and Item 3 of the 2025 10-K could not be read; Item 1C was re-read in the primary text
  • Independent testing or red-teaming of the AI customer service agent
  • Error rates of reservation screening and other automated means
  • Official statement of the foundation models in use
  • Tension between the denial of automated decisions and the consequences of reservation screening up to account removal; not legally assessed

7 · How it was assessed

Cut-off date: 3 November 2026 · 41 searches, 46 fetches.

Primary documents read (11)
  • Form 10-K 2025 (partial, through Item 1A payment risks)
  • Proxy Statement DEF 14A 2026 (partial, up to CD&A)
  • DPC Decision Airbnb Ireland UC of 2024-01-31
  • EDPB/DPC Decision No 1364 (amicable resolution, 2024-07-05)
  • Airbnb Ireland UC DSA Transparency Report (2025-02-14)
  • Sustainability and Community Update 2023
  • Project Lighthouse Update 2024
  • Law Enforcement Response Transparency Report 2024
  • Help Centre articles 4097, 4178, 3280, 2863, 3724
  • Q4 2025 earnings call transcript
  • Form 10-K 2025, Item 1C Cybersecurity (XBRL R28) and Commitments and Contingencies – Narrative (XBRL R72), review dated 2026-11-03

Research of 3 November 2026 under Globalance AI Footprint rulebook version 0

We assess documented practice in dealing with AI; this is not an investment recommendation. Three pillars, weights 40/40/20, harm rule.

Research of 3 November 2026 under Globalance AI Footprint rulebook version 0. Documented observations from public sources, no investment recommendation.

Important Notice. This material is provided for information purposes only and does not constitute an offer or solicitation to buy or sell any financial instrument, nor does it constitute investment advice or a recommendation. The GI 100 Index is a rules-based strategy index of Globalance Bank. Investments involve risk, including the loss of capital invested. The final product documents (including FinSA) shall prevail in all cases.

Your company? Information on inclusion in the GI 100 Index →

What comes next