ABNB · Airbnb, Inc.

conf 3/10 partial

enthusiasm:27.0 · trend:8 · quantifies:5 · impact:0 · under_radar:0 · credibility:12 · business_impact:8 · disruption:-6 · commitment:6 · confirmation:3

Enthusiasm latest 9 / prev 8 (rising)

Airbnb's AI thesis is now both product-facing and operating-model-facing: customer support automation, code generation, search/relevance, personalization, host tools and listing creation. Credibility improved in Q1 FY2026 because management tied AI to concrete realized metrics including over 40% support self-solve, nearly 60% AI-authored code and a roughly 10% year-over-year decrease in cost per booking. Management still says AI search has no top-line benefit baked into guidance, so the proven impact is mostly productivity and support cost leverage rather than revenue acceleration.

Grounded on actual base — revenue $12.2B · net income $2.5B · net margin 20.5% · diluted EPS 4.03

These are next-fiscal-year annual uplift estimates, not next-quarter numbers.

Aggregate next-FY est. rev uplift: 0.0% · next-FY EPS uplift: % · vs analysts: unclear · priced in: high (model's call-read: medium; verdict above is the hard-data one used for ranking) · confidence: 3/10

Assumptions: Tax rate 21% on any cost saving; incremental net margin = company net margin 20.51% if any topline existed (none disclosed). Phasing: management frames the AI support build as an EXPENSE that RAMPS over 2026, so near-term it is cost-neutral-to-negative before efficiencies compound. Segment mapping: all claims are own-operations (support, engineering) — none sell AI capacity. No support-cost or bookings dollar base is disclosed in any claim, so cost-saving claims cannot be dollarized without inventing a base (prohibited).

Top line: Effectively zero for the next fiscal year by management's own framing: 'We have nothing baked into our outlook in terms of the benefit from that deployment.' AI trip-planning/search is real optionality but is explicitly excluded from guidance and carries no quantified figure, so the disciplined next-FY revenue uplift is 0%. Consensus FY26 revenue of $13.955B (+14.0% over $12.241B base) reflects core growth, not an AI topline.

Bottom line: Directionally positive but not dollar-anchored. The hardest signals are cost per booking down ~10% YoY (Q1'26) and support deflection rising from ~33% (Q4'25) to >40% (Q1'26) — both lower live-agent cost. However, no support-opex or bookings base is disclosed anywhere (implied COGS is only $2,086M = $12,241M rev − $10,155M gross profit, and support is just a fraction of that and not isolable), so an after-tax-saving / $2,511M NI EPS uplift cannot be computed without inventing a base — left null per guardrail. Offsetting near-term: management flags the AI customer-service buildout as an expense that 'will ramp over the course of the year.'

Consensus already embeds margin/EPS expansion: FY26 EPS $5.128 vs FY25 base $4.03 (+27.2%) and net income $3,097M vs $2,511M (+23.3%) on +14% revenue — i.e. net margin rising ~20.5% → ~22.2%, which plausibly absorbs the ~10% cost-per-booking efficiency into the existing trajectory (priced in). The genuinely un-priced piece — AI trip-planning topline — is explicitly NOT in the outlook, but management attaches no number to it, so the gap cannot be quantified against the $13.955B/$5.128 consensus. Net: cost side ≈ priced in; topline side = un-priced but unsized.

• We built a custom AI agent trained on millions of our support interactions. It's already resolving 1/3 of the support issues without needing a live specialist and resolution times are significantly faster.
• Right now, nearly 30% of tickets in North America that are English-based are handled by an AI agent.
• We've seen the cost per booking decrease about 10% year-over-year in Q1, and we expect to see more of this as we improve AI customer support this year.

• Nearly 60% of the code our engineers produce is now written by AI, which we estimate is about twice the industry average.
• You see when guests contact us through our AI assistant, over 40% of issues are now resolved without a human agent.
• We're now using AI for matching. AI is really helping our search ranking and our relevance.
• Actually, funny enough, we are doing tests as we speak. So AI search is live to a very small percent of traffic right now.

• It will help guests plan their entire trip, help us better run their businesses and help the company operate more efficiently at scale.
• A year from now, if we're successful, significantly more than 30% of tickets will be handled by a customer service agent in many more languages in all the languages where we have live agents and AI customer service will not only be chat, it will be voice.
• We have nothing baked into our outlook in terms of the benefit from that deployment.
• And so I would anticipate that, that is an expense that will ramp over the course of the year.

70/100 track record   delivers  6 calls reviewed

Airbnb makes few hard-dated AI commitments, and many disclosed metrics are reported outcomes rather than prior promises; but the clearest dated commitment — AI customer support by the 2025 summer release — was delivered on time and then exceeded (15% fewer human contacts, then 1/3, then 40%+ resolution, ~10% lower cost per booking). Remaining targets (50 languages, conversational search, 30% eng productivity) are too-early.

Est. revisions rising  ·  Fwd P/E 32.5  ·  EV/Sales 5.9x

AI claim maps to Reportable Segment

AI enthusiasm across 6 calls — trend ↗ rising

AI moved from vague search/product optimization to quantified support automation, AI-native search/personalization, engineering productivity, and cost savings.

• 2026-05-21 · Airbnb launches major expansion with airport pickups, luggage storage and AI-powered travel tools
• 2026-05-20 · Airbnb gets into hotels, expands AI for host onboarding and customer support
• 2026-05-20 · Airbnb is adding AI, hotel bookings, and social features in its biggest app overhaul in years
• 2026-05-20 · How Airbnb CEO Brian Chesky Learned to Stop Hating Hotels
• 2026-05-08 · Airbnb CEO Predicts AI Will Rewrite Job Descriptions
• 2026-05-08 · Airbnb says AI now writes 60% of its new code

8/10 qualitative impact   material  medium-term · mixed evidence

Where AI matters: customer support, search, personalization, host tools and engineering productivity

Airbnb has real deployed AI with quantified operating impact: >40% support issues resolved without a human, AI-written code near 60%, and cost per booking down ~10% YoY. The revenue upside from AI search/trip planning and host onboarding could be meaningful, but management says no top-line benefit is baked into the outlook, so the transformational case remains unproven.

Caveats: AI search/trip-planning adoption may not convert into incremental bookings; Support automation savings may be offset by AI buildout expense; Third-party AI assistants could intermediate travel discovery and reduce platform control; Regulatory, trust and quality issues still constrain automation in lodging marketplaces

AI DISRUPTION / CANNIBALIZATION RISK  two-sided · 4/10

AI travel agents and conversational search could weaken Airbnb's control of discovery, raise aggregation pressure, and commoditize parts of trip planning, but the core model is anchored by unique supply, trust, payments, reviews and host/guest network effects that are hard to automate away. The bigger risk is front-end distribution and take-rate pressure, not existential replacement of lodging inventory.

option liquidity: good

proxy inputs — dollar-ADV $498M · beta 1.208 · px $134.35

source: proxy (no options chain on FMP)
FMP /stable/ exposes no options-chain endpoint on this key, so ATM IV, bid-ask spread and open interest are unavailable. Liquidity below is a PROXY from dollar-ADV, beta and price level (a stand-in for option depth), not measured option-market data.