Brian Nowak

Brian Nowak asked about the long-term revenue and ROIC opportunities from Meta's AI and MetaCompute investments over the next 3-10 years, and the key drivers of the accelerated revenue growth expected in 2026.

Answer

CEO Mark Zuckerberg outlined long-term opportunities including improving core products with AI/LLMs, new business models like Meta AI monetization (subscriptions, advertising), and enhanced shopping/commerce tools. CFO Susan Li attributed the strong Q1 2026 revenue outlook to robust demand, significant ad performance improvements (ranking, delivery, ad load redistribution, Advantage+, better measurement), and a 4% foreign currency tailwind.

Brian Nowak asked about Meta's long-term revenue and ROIC opportunities from AI and Meta Compute over the next 3-10 years, and the biggest drivers behind the expected revenue inflection in 2026.

Answer

CEO Mark Zuckerberg discussed long-term business opportunities including improving core products through AI-driven recommendation systems, new media generation, Meta AI monetization via subscriptions and advertising, and growth in shopping/commerce, including the Manus acquisition. CFO Susan Li attributed the 2026 revenue inflection to strong demand, a significant foreign currency tailwind, and improved ad performance from investments in ranking, delivery, ad load redistribution, and new ad products.

Brian Nowak (Morgan Stanley) inquired about the quantifiable signals and A/B test results that instill confidence in the return on invested capital (ROIC) from Meta's significant 2026 infrastructure build, particularly for core AI and ad ranking model improvements. He also asked for the estimated Reality Labs revenue headwind in the Q4 guidance.

Answer

CFO Susan Li explained that 2026 CapEx growth is driven by Meta Superintelligence Labs, core AI, and non-AI, with AI needs growing most. She highlighted strong year-over-year growth in value-weighted conversion rates, exceeding impression growth, as a key performance indicator for ads. Susan Li also noted that the new model architecture (Lattice) is enabling better ad performance with more data and compute. Regarding Reality Labs, she stated that Q4 revenue is expected to be lower due to lapping the Quest 3S launch in Q4 last year and early holiday procurement of Quest headsets in Q3 2025, despite strong AI glasses revenue.

Brian Nowak asked about quantifiable signals for ROI from 2026 CapEx investments in AI and ad ranking models, and the expected Reality Labs revenue headwind in Q4 guidance.

Answer

CFO Susan Li explained that 2026 CapEx growth is across MSL, core AI, and non-AI, with MSL AI growing most. She highlighted strong year-over-year growth in value-weighted conversion rates and the impact of new model architectures on ad performance. Regarding Reality Labs, she noted Q4 revenue would be lower due to lapping Quest 3S launch in Q4 2024 and Q3 2025 sales recognition, partially offset by strong AI glasses demand.

Brian Nowak of Morgan Stanley inquired about the primary technological gating factors for achieving superintelligence that Meta is focused on overcoming in the next 24 months. He also asked about the most exciting upcoming drivers for engagement on Meta's core platforms.

Answer

CEO Mark Zuckerberg identified AI self-improvement as a fundamental research area, noting that superintelligence must learn beyond human input. He also highlighted a shift toward using small, talent-dense teams for frontier research. CFO Susan Li outlined near-term engagement drivers like more adaptive recommendations and better content discovery for smaller creators, with long-term bets on foundational models and deeper LLM integration.

Brian Nowak inquired about the primary technological gating factors for achieving superintelligence and asked for the key drivers of future user engagement on Meta's core platforms.

Answer

CEO Mark Zuckerberg identified AI self-improvement as a fundamental research area and noted that small, talent-dense teams are optimal for frontier research. CFO Susan Li outlined near-term engagement drivers like more adaptive recommendations and better content discovery, and long-term bets on foundational models and deeper LLM integration.

Brian Nowak of Morgan Stanley asked about the evolution of technological constraints in the pursuit of superintelligence over the last year. He also inquired about the most exciting upcoming factors expected to drive further user engagement on Meta's core platforms.

Answer

CEO Mark Zuckerberg identified self-improvement as a key research area for superintelligence, noting that AI must learn to improve itself beyond human-provided data. He also emphasized the effectiveness of small, talent-dense teams for frontier research. CFO Susan Li outlined several engagement drivers, including making recommendations more adaptive within a session, helping content from smaller creators surface faster, and longer-term bets on foundational models and LLM integration.

Brian Nowak inquired about Meta's roadmap for its Llama large language models, specifically the key areas of advancement for future versions like 'Behemoth', and asked for details on the user traction and recurring behaviors seen with the Meta AI product.

Answer

CEO Mark Zuckerberg explained that Llama's development is focused on optimizing for Meta's infrastructure and use cases, such as low-latency voice conversations and long context windows for personalization. He noted that large models like 'Behemoth' are used to distill more efficient production models. CFO Susan Li added that the top use case for Meta AI is information gathering, followed by social interactions, with WhatsApp being the platform with the strongest usage.

Brian Nowak of Morgan Stanley inquired about Meta's vision for new use cases driven by Llama 4 and Meta AI in 2025, and for insights into the performance and adoption timeline for its custom silicon (MTIA) versus third-party chips.

Answer

CEO Mark Zuckerberg highlighted the focus on Meta AI as a personalized assistant and the goal for Llama 4 to be a leading, multimodal open-source model, while keeping specific 2025 product launches a surprise. CFO Susan Li explained that custom silicon (MTIA) is optimized for Meta's unique workloads to improve cost and power efficiency, noting it's currently being ramped up for ranking and recommendation inference with plans to expand to training workloads next year.

Brian Nowak inquired about the recurring user interactions and commercial intent within Meta AI, the strategy for in-house search versus partnerships, and the outlook for headcount investment in 2025.

Answer

CFO Susan Li explained that Meta AI sees frequent use for information gathering, content discovery, and image generation, with potential for more monetizable queries over time. She noted that Meta AI integrates with search partners like Bing and Google. Regarding 2025 headcount, Li stated that budgeting is ongoing, with a focus on investing in strategic priorities like monetization and Gen AI while maintaining operational discipline.

Brian Nowak asked for an update on the progress of Uber's urban versus suburban strategy within the U.S. mobility business. He also sought insight into key investment areas for the European food delivery business into 2026, considering increased competitive activity.

Answer

CFO Prashanth Mahendra-Rajah explained that sparse geographies are growing 1.5 times faster than denser markets, presenting significant penetration opportunities driven by product availability, reliability, and tailored offerings like Wait & Save. CEO Dara Khosrowshahi affirmed Uber's leading position in Europe, emphasizing organic growth, expanded merchant selection, service improvements, cross-sell, membership, and strategic partnerships to navigate competition.

Brian Nowak asked for more color on the U.S. mobility business's performance, specifically regarding the urban versus suburban strategy, and sought details on Uber's European food delivery business in light of increased competition and key investment areas for 2026.

Answer

CFO Prashanth Mahendra-Rajah explained that sparse U.S. geographies are growing at 1.5 times the rate of denser markets, with significant penetration opportunity, driven by product availability, reliability, and fit (e.g., Wait & Save). CEO Dara Khosrowshahi stated that Uber holds a leading position in Europe, gaining market share in key countries like Spain and Germany, and plans to continue focusing on merchant selection, service improvement, cross-sell, and membership despite competition.

Brian Nowak inquired about the specific factors driving the strong sequential growth in Monthly Active Platform Consumers (MAPCs) and Uber One members, and asked about the durability of this growth. He also requested an update on the scale of autonomous vehicle (AV) rides and Waymo utilization on the Uber network.

Answer

CEO Dara Khosrowshahi attributed the 15% audience growth to a strategy of introducing lower-cost products like Moto and expanding premium offerings. He emphasized significant runway remains, as only 20% of the addressable population in Uber's top 10 markets are monthly users. For Uber One's 60% growth, he pointed to new features like 'surge savings' for Mobility users as a key driver. Regarding AVs, he noted that the average Waymo vehicle on the platform is busier than 99% of human drivers in its operating zones, indicating strong utilization and a positive 'halo effect' on the brand.

Brian Nowak asked for a timeline on when the Waymo fleet in Austin might be large enough to drive incremental volume for Uber and for an update on U.S. mobility growth, particularly in suburban versus urban areas.

Answer

CEO Dara Khosrowshahi stated the immediate goal in Austin is perfecting the user experience and safety, not yet driving incremental trip volume. CFO Prashanth Mahendra-Rajah added that overall mobility trip growth remains strong and that sparser, suburban markets now constitute about 20% of mobility trips and are growing faster than the urban core.

Brian Nowak of Morgan Stanley inquired about Uber's investment philosophy in autonomous vehicle (AV) assets and the expected trajectory of incremental margins for the core Rides business in 2025.

Answer

CEO Dara Khosrowshahi stated that Uber is investing aggressively across the AV portfolio, including in fleet supply and partnerships, but these investments are not expected to impact the company's three-year financial outlook. CFO Prashanth Mahendra-Rajah added that the company makes investment decisions at a total company level and will continue to invest in Mobility to drive growth, highlighting strong performance drivers like the Uber One membership program and Uber Teens.

Brian Nowak asked for an update on U.S. Mobility booking trends relative to prior outlooks and inquired about any early signs of incremental volume from the Waymo partnership in Arizona.

Answer

CEO Dara Khosrowshahi noted that U.S. Mobility remains strong, though transaction growth has slowed due to the pass-through of higher insurance costs, consistent with expected price elasticity. He highlighted strong weekday and Uber for Business growth. Regarding the Waymo partnership, Khosrowshahi stated it's too early to measure incrementality in Phoenix but emphasized the positive user experience. He indicated the upcoming expansions in Austin and Atlanta will be a more significant test.

Brian Nowak inquired about the strategic positioning of Amazon's Trainium chips versus third-party alternatives and the key challenges Amazon needs to overcome to broaden Trainium's adoption, rather than relying solely on third-party chips for demand.

Answer

CEO Andy Jassy emphasized AWS's commitment to offering multiple chip options, maintaining deep relationships with partners like NVIDIA, and continuing significant purchases. He highlighted Amazon's strong custom silicon team (Annapurna) and the price-performance advantage of Graviton (CPU) and Trainium (AI) chips. Jassy stated that Trainium3 is expected to be 40% better than Trainium2 in price-performance, and key hurdles include delivering the chip in volume quickly, enhancing the software ecosystem, and building credibility through successful large-scale projects like Project Rainier.

Brian Nowak inquired about the strategic positioning of Trainium versus third-party chips and the key challenges AWS needs to overcome to broaden Trainium 3 adoption.

Answer

CEO Andy Jassy emphasized AWS's commitment to offering multiple chip options, maintaining a deep relationship with NVIDIA, and continuing significant NVIDIA purchases. He highlighted Amazon's strong custom silicon team (Annapurna) and the price-performance advantage of Graviton (40% better than x86) and Trainium (similar benefit over GPUs). Jassy expects Trainium 3 to be 40% better than Trainium 2, focusing on delivering the chip in volume, quickly, and enhancing the software ecosystem, with Project Rainier serving as a key proof point.

Brian Nowak of Morgan Stanley asked for a rebuttal to the narrative that AWS is falling behind in GenAI and questioned if AWS revenue growth should be expected to accelerate in the coming year.

Answer

CEO Andy Jassy countered the narrative by emphasizing it's still 'very early' in AI and detailed AWS's full-stack advantage, from custom Trainium chips to comprehensive services like Bedrock and Agent Core. While not providing guidance, Jassy expressed optimism for AWS acceleration, citing resumed enterprise cloud migrations, growing AI adoption, and increasing supply capacity as key positive drivers.

Brian Nowak asked for a rebuttal to the narrative that AWS is falling behind in Generative AI and questioned if there's any reason AWS growth shouldn't accelerate into late 2025 and 2026.

Answer

CEO Andy Jassy countered the 'falling behind' narrative by emphasizing that it's still very early in AI. He detailed AWS's comprehensive stack, from custom Trainium chips to services like Bedrock and Agent Core, and the strategic advantage of customers' data already residing on AWS. While not providing specific guidance, he expressed optimism for future growth driven by enterprise cloud migration and scaled AI adoption.

Brian Nowak asked about the key operational areas of focus for the retail business to ensure a smooth holiday season amidst tariff uncertainty and if the Q2 EBIT guide contains one-time costs.

Answer

CEO Andy Jassy highlighted managing inventory levels within the fulfillment network as a key focus to handle potential early inventory inflows from sellers without compromising productivity. CFO Brian Olsavsky confirmed the Q2 guide's wider range reflects general uncertainty, but its primary cost drivers are seasonal stock-based compensation and Kiper expenses, not a specific one-time tariff charge like in Q1.

Brian Nowak requested an update on learnings from the Shreveport robotics facility and its scalability, and also asked for more examples of how generative AI will change the Amazon retail experience in 2025.

Answer

CEO Andy Jassy expressed strong encouragement from the Shreveport facility, which integrates a new tranche of robotics, citing improvements in speed, productivity, and cost to serve, with plans for a broader rollout. He detailed how AI is already enhancing the retail business through productivity tools (e.g., customer service chatbots, seller listing tools) and new customer experiences like the Rufus shopping assistant, Amazon Lens, and sizing recommendations, noting about 1,000 gen AI apps are in development.

Brian Nowak inquired about the drivers of international retail profitability and its long-term margin potential, and also asked for an update on the company's robotics investment journey.

Answer

CFO Brian Olsavsky stated that international profitability is driven by the same factors as North America, including lower cost-to-serve and advertising, with the long-term goal of reaching North America-level margins. He also noted that while Amazon leads in robotics, it's still 'early days,' highlighting a new facility in Louisiana that integrates the latest automation to improve speed and lower costs.

Brian Nowak from Morgan Stanley asked why Pinterest's AI and machine learning improvements haven't led to a significant step-up in growth acceleration, as seen on other platforms, and whether there are any underlying constraints.

Answer

CEO Bill Ready reiterated his consistent message that these are long-term, compounding initiatives, not 'hockey stick' or 'step function' changes. He emphasized that Pinterest is consistently growing above the market rate by transforming into a performance ad platform, taking share, and sees no structural limitations to this steady, compounding growth.

Brian Nowak of Morgan Stanley asked why Pinterest's AI improvements haven't led to a more significant step-up in advertising revenue growth, questioning if there are underlying constraints on acceleration.

Answer

CEO Bill Ready rejected the idea of a 'step function,' reiterating that these are long-term, compounding initiatives. He emphasized that Pinterest is consistently growing above the market rate by transforming into a performance ad platform and taking share. He pointed to steady progress in the mid-market and international regions as evidence of this compounding growth without structural limitations.

Brian Nowak asked for the specific lift in advertiser spend from adopting Performance+ and inquired about the strategy to close the gap between rapid click growth and ad revenue growth.

Answer

CEO Bill Ready did not provide a specific percentage lift for Performance+ but noted that 80% of A/B tests with the suite outperform traditional campaigns. He explained the gap between click and revenue growth is closing as advertisers adopt new measurement tools (like CAPI) and campaign automation, which validates performance and unlocks budget.

An analyst on behalf of Brian Nowak at Morgan Stanley inquired about the most significant potential drivers for platform improvement from GPU-enabled machine learning and Gen AI in 2025.

Answer

CEO Bill Ready emphasized that AI is a core competency, powered by Pinterest's unique 'Taste Graph' which leverages hundreds of billions of user actions. He highlighted using GPUs for larger models, LLM technology for guided search, and computer vision for features like 'Shop the Look.' He also noted significant internal productivity gains, with AI-assisted tools generating roughly 15% of the company's current code base.

Brian Nowak asked about Pinterest's most important ad innovation priorities for 2025, the influence of generative AI, and the potential to achieve durable revenue growth in the 20% range.

Answer

CEO Bill Ready highlighted that the key growth levers for 2025 are the continued build-out of the full-funnel ad platform, particularly lower-funnel tools and the new Performance Plus automation suite. He noted that excluding the challenged food and beverage vertical, Pinterest's revenue growth was already in the low 20% range in Q3, demonstrating the underlying strength and durability of its performance advertising business.

Brian Nowak asked about the areas where improving the product, customer experience, and merchant services has proven to be more challenging than initially expected.

Answer

CEO & Chair Tony Xu countered the premise of the question, stating that based on the hundreds of customer emails he reads weekly, the company is falling short and has opportunities for improvement everywhere. He gave an example of a delivery to the wrong parking lot, illustrating that even small-sounding issues are significant at scale. He concluded that there is no single area that is harder to crack; rather, every day is a struggle with opportunities for improvement across the entire platform, from minor execution details to major strategic shifts with AI.

Brian Nowak asked for examples of lessons learned from Wolt that give DoorDash confidence in acquiring Deliveroo, which is a market laggard in some areas. He also asked about the drivers behind the significant increase in CapEx and the outlook for the year.

Answer

CEO Tony Xu explained that Deliveroo has built leading positions in strong profit pools and that lessons from the Wolt integration, such as exchanging product knowledge, provide a proven playbook. CFO Ravi Inukonda attributed the CapEx increase to a merchant tablet refresh and investments in autonomy, stating the full-year level should be in line with the overall D&A guidance.

Brian Nowak asked how the company's investment priorities and capital allocation for 2025 have changed compared to 2024, and also inquired about the expected pace of the share buyback program.

Answer

CEO Tony Xu stated that investment priorities have not changed significantly, as he continues to see strong reasons to invest across all key business areas. CFO Ravi Inukonda added that the capital allocation philosophy is also unchanged, with a focus on generating returns, and that the company will continue to be opportunistic with its share buyback program.

Brian Nowak asked for the key priorities to accelerate room night growth in the U.S. and EMEA into 2026, noting that growth in EMEA trails a scaled competitor.

Answer

CEO Brian Chesky outlined a multi-pillar strategy focused on the U.S. market. The key priorities include improving ease of use, enhancing pricing and affordability through tools like total price display and better host pricing suggestions, and strategically increasing supply. This supply effort involves targeting the most constrained markets and onboarding more hotels to supplement home availability.

Brian Nowak asked about the key points of friction that Airbnb's new products and services aim to address for guests and hosts, and requested a breakdown of the drivers behind the full-year margin guidance.

Answer

CEO Brian Chesky outlined a philosophy of expanding beyond stays into adjacent travel services, aiming to make the core business stronger and increase app usage frequency. CFO Ellie Mertz detailed the margin guide, explaining that the $200-$250M investment will impact margins most in Q1-Q3, with revenue benefits concentrated in Q4, while highlighting the calendar and FX headwinds affecting the Q1 outlook.

Brian Nowak questioned the drivers behind the Q4 Adjusted EBITDA margin guidance, which implies some compression. He also asked about the company's broader philosophy on investment levels and margins heading into 2025.

Answer

CFO Ellie Mertz attributed the Q4 margin outlook to timing shifts and incremental investments in product development and marketing, including for global expansion. Looking to 2025, she stated that while there is long-term opportunity for margin expansion, the company will continue to lean into growth initiatives for its core business and new offerings. These investments will be capital-light but will include hiring and marketing spend to support growth levers.

Brian Nowak of Morgan Stanley inquired about U.S. user trends during Q2 and into Q3, and asked which advertising innovations management is most excited about for the second half of the year to drive return on ad spend (ROAS).

Answer

CEO Steve Huffman acknowledged a slow start to Q2 user growth in April but noted improved traction later in the quarter, with the exit rate higher than the 110M average and continued growth in July. He mentioned Google traffic was a headwind. COO Jen Wong highlighted excitement for the broader rollout of Dynamic Product Ads (DPA), increased platform automation, and scaling new solutions derived from Reddit's Community Intelligence, such as Reddit Insights and Conversation Summary add-ons.

Brian Nowak of Morgan Stanley inquired about U.S. user trends during Q2 and into Q3, and asked about the key advertising innovations planned for the second half of the year to drive return on ad spend (ROAS).

Answer

CEO Steve Huffman confirmed that after a slow start in April, U.S. user trends improved through Q2 and into July, driven by product and marketing efforts. COO Jen Wong highlighted upcoming ad innovations, including the broader rollout of Dynamic Product Ads (DPA), increased platform automation, and new 'Community Intelligence' solutions like Reddit Insights, which are designed to boost advertiser performance.

Brian Nowak from Morgan Stanley asked about the key initiatives for converting logged-out users to logged-in users and which advertising product improvements represent the lowest-hanging fruit for 2025 versus those that are longer-term projects.

Answer

CEO Steve Huffman detailed that for direct visitors, simplifying the sign-up process and improving onboarding have been key. For users from search, the focus is on providing great answers first to build trust. COO Jen Wong identified applying ML optimization to brand advertising, improving measurement, and creative optimization as lower-hanging fruit for the ad business, with continuous improvement in ease-of-use and unique formats being ongoing efforts.

Brian Nowak of Morgan Stanley asked for a walkthrough of the key ad innovations expected to drive growth in 2025 and for a conceptual timeline on building and monetizing on-platform search.

Answer

COO Jen Wong detailed the 2025 ad strategy, focusing on enhancing brand advertising with more automation, building out lower-funnel shopping capabilities, and creating end-to-end automation. CEO Steve Huffman added that search development will be a focus in 2025, with monetization to follow once the product is stable, suggesting revenue impact would likely be after 2025.

Brian Nowak asked for a long-term vision on how Generative AI will evolve Google Search and for specific examples of where Alphabet can drive further operating expense efficiencies.

Answer

CEO Sundar Pichai explained that AI will expand Search's capabilities into more powerful, multimodal, and agentic use cases, citing Project Astra and Gemini Deep Research as examples. CFO Anat Ashkenazi identified key areas for efficiency gains, including optimizing technical infrastructure, moderating headcount growth, reducing the real estate footprint, and simplifying the organization by using internal AI tools.

Brian Nowak inquired about the current macro advertising environment, asking for any signs of weakness or changes from typical seasonality, and also questioned which products are driving the increase in commercial search queries.

Answer

Philipp Schindler, Senior Vice President and Chief Business Officer, responded that Q1 saw broad-based strength in ad verticals, led by finance, retail, and healthcare. He noted it was too early to comment on Q2 but mentioned a slight headwind from changes to the de minimis exemption. Schindler identified AI Overviews as the core driver for the increase in commercial queries, stating the company is confident in its performance and potential for expansion.

Brian Nowak inquired about the most promising generative AI-enabled search products for driving long-term growth and the primary constraints to scaling them globally.

Answer

CEO Sundar Pichai highlighted AI Overviews, Circle to Search, and Lens as key innovations expanding search capabilities and driving user engagement. He noted that while constraints like latency and cost per query exist, Alphabet has made substantial progress, reducing machine costs by over 90% in 18 months, and does not see them as a major barrier to scaling.

Brian Nowak inquired about the traction of generative AI over the past 18 months, asking which areas have seen faster or slower adoption than expected, and requested tangible examples of ongoing cost-cutting initiatives.

Answer

CEO Sundar Pichai noted that consumer-facing AI in products like Search has shown positive traction, while enterprise solutions still require more work to unlock their full value. CFO Ruth Porat emphasized the company's focus on "durably reengineering our cost base" through product prioritization, organizational efficiency like combining teams, and optimizing technical infrastructure, procurement, and real estate.

Brian Nowak of Morgan Stanley asked about specific internal growth initiatives aimed at accelerating U.S. market performance and the primary technological hurdles to deploying scalable GenAI travel assistants.

Answer

CFO Ewout Steenbergen explained the U.S. strategy involves numerous small initiatives across product, supply, and marketing to achieve gradual market share gains, noting Booking likely outpaced the industry in Q2. CEO Glenn Fogel described the GenAI challenge as an incremental process of building a superior, personalized service that integrates proprietary data to proactively solve travel issues, with features like natural language search being an early step in this evolution.

Brian Nowak of Morgan Stanley asked about specific internal growth initiatives to accelerate U.S. market performance and the primary technological hurdles for deploying scalable GenAI travel assistants to a mass audience.

Answer

CFO Ewout Steenbergen explained the U.S. strategy involves numerous small initiatives across product, supply, and marketing that are contributing to market share gains, despite mixed signals like lower ADRs. CEO Glenn Fogel addressed GenAI, stating the challenge is an incremental process of integrating proprietary data and models to create a superior, personalized service that drives direct traffic, a process already underway with features like natural language search.

Brian Nowak questioned the decision to widen the full-year guidance range despite commentary on stable business trends and asked for more specifics on the lower-ROI marketing spend driving growth.

Answer

CFO Ewout Steenbergen explained that the guidance range was widened to account for increased geopolitical and macroeconomic uncertainty for the second half of the year, while the high end of the range remains unchanged. Regarding marketing, he gave social media channels as an example where scaling up provides attractive, positive ROIs that are accretive to the business, even if they are below the average ROI of more established channels.

Brian Nowak asked how Booking Holdings manages the long-term risk of AI partners using its supply to build competing consumer-facing products, potentially harming Booking's direct traffic mix.

Answer

CEO Glenn Fogel explained that partnership agreements are carefully structured to manage data and economics, emphasizing Booking's strong negotiating position due to its brand trust and data. CFO Ewout Steenbergen added that a proliferation of AI agents could be an economic advantage, creating more customer acquisition channels and potentially lowering overall acquisition costs for Booking.

Asked about the primary technological hurdles for launching scalable commercial agents and requested an update on internal efficiencies gained from GenAI capabilities.

Answer

Sundar Pichai identified reliability, latency, and cost as the key challenges for agentic AI, but noted strong progress and expects broader use in 2026. He also confirmed that Google is already deploying agentic coding tools internally to improve software engineering efficiency, overcoming initial friction points.

Brian Nowak asked about the 2025 budgeting process, including any constraints on faster top-line growth and the trade-offs between growth and profitability. He also asked for the size of the restaurant business as a percentage of GTV.

Answer

CFO Emily Maher described a dynamic budgeting process focused on reinvesting efficiency gains into high-return growth initiatives while maintaining a commitment to annual EBITDA expansion. She declined to break out the restaurant GTV percentage, emphasizing its flywheel effect on the core grocery business and noting that adoption is still in its early stages.

Brian Nowak asked about the flat sequential shopper growth and the keys to accelerating it, and also inquired about the top ad innovations that could drive ad revenue to outpace GTV growth in 2025.

Answer

CEO Fidji Simo clarified that shopper supply is very healthy, with waitlists in many cities, and the focus is on efficiency through batching and optimized routing, not just growing the total number of shoppers. For advertising, she identified three key drivers for 2025: proving ad performance with superior measurement, launching new formats like shoppable recipes, and expanding ad supply through Carrot Ads on retailer sites and Caper screens.

Brian Nowak asked for details on the go-to-market strategy for the 'Price Lock' feature and inquired about any observable trends in San Francisco volumes due to AV competition.

Answer

CEO David Risher stated Price Lock is rolled out nationally, targeting commuters, and is showing low churn and driving incremental rides. Regarding San Francisco, he noted that while AVs are visible, Lyft views them as future partners and that current AV usage is skewed by novelty and tourism, making it too early to see a competitive impact on volume.