A：In Meta's third quarter, the key achievements and milestones include reaching a major milestone with 3 billion monthly active users on Instagram, good momentum across other apps likeThreads, with 150 million daily actives and on track to become the leader in its category.

A：Meta's vision is to establish itself as the leading frontier AI lab, aiming to build personal superintelligence for everyone and deliver app experiences and computing devices that improve lives worldwide. Meta's approach involves advancing open source AI, developing next-generation models and products, and increasing compute capacity.

A：MetaSuperIntelligences Labs has started strongly, developing with high talent density. Meta is focused on building next-generation models and products and is investing in industry-leading compute capacity.

A：People have various timelines for achieving superintelligence, ranging from a few years to longer than 7 years. Meta's strategy is to aggressively front-load building capacity to prepare for the most optimistic cases and use extra compute to accelerate the core business in case of a delay.

A：Meta's ads business is performing very well, driven by improvements in AI ranking systems. This quarter saw meaningful advances from unifying different models into simpler, more general models, increasing both performance and efficiency.

A：Meta is excited about new products such as Meta AI, with over a billion monthly actives, and usage increases with improved underlying models. The company is also excited about business AI, with over 1 billion active threads, helping businesses scale conversations and improve sales at low cost.

A：Meta has launched Vibes, the next generation of AI creation tools and content experiences, with growing weekly usage. Meta is optimistic about ramping up growth in the coming months.

A：The response to Meta's new line of AI glasses has been great, with products selling well and improved battery life, camera resolution, and AI capabilities. Meta plans to invest in increasing manufacturing to meet the demand.

A：Meta's overall financial performance in the third quarter includes total family of apps revenue up 26% year over year, and Q3 revenue of $51.2 billion, up 26% or 25% on a constant currency basis. Expenses grew 32% compared to last year, driven by legal, employee compensation, and infrastructure costs. Meta ended the quarter with over 78,400 employees, up 8% year over year.

A：The new US tax law resulted in a one-time non-cash reduction in deferred tax assets, which unfavorably impacted the company's effective tax rate for the quarter. Excluding this charge, the tax rate would have been 14%. The company continues to expect significant cash tax savings for the remainder of the year and future years under the new law, and the third quarter's charge reflects the total expected impact from this transition.

A：The company's capital expenditures in the third quarter were driven by investments in servers, data centers, and network infrastructure.

A：The two primary factors that drive the company's revenue performance are its ability to deliver engaging experiences for its community and its effectiveness at monetizing that engagement over time.

A：The projected impact of improved recommendation systems is to continue the year-over-year growth in global time spent on both Facebook and Instagram in the US, with overall time spent growing double digits, driven by continued video strength and healthy growth in non-provided content over time.

A：The focus areas for the recommendation systems on Instagram in 2026 include evolving the systems to surface content across a broader set of topics to cater to the diverse interests of each person, making significant progress on long-term ranking innovations, and leveraging language models to improve content understanding.

A：The number of people using Meta AI across the company's family of apps is growing, and the company is increasingly leveraging first-party content into Meta AI results. Meta AI responses to Facebook deep dive queries in the US are now showing related reels, and there has been a more than tenfold increase in media generation within the app since launching AI-generated imagery in September.

A：The company is optimizing the level of ads within organic engagement by refining the supply across each of its major services to better deliver ads at the most relevant times for people. In the long term, they plan to globally introduce ads on WhatsApp status, following their typical monetization playbook of optimizing ad formats and performance before ramping supply.

A：Improvements to marketing performance and ad systems include ongoing enhancements to large-scale ads ranking models, the adoption of Lattice, a unified model architecture, which has driven gains in conversions. The company has also been consolidating smaller and more specialized models into larger ones using the Lattice architecture, leading to performance improvements.

A：The company expects total revenue for the fourth quarter of 2025 to be in the range of $56 to $59 billion. The guidance assumes foreign currency is an approximately 1% tailwind to year-over-year total revenue growth based on current exchange rates.

A：The company plans to invest in ads and organic engagement initiatives, as well as AI models and products, to enable strong revenue growth in 2026. They expect a significant increase in CapEx dollar growth in 2026, with total expenses growing at a faster rate due to infrastructure costs and employee compensation. The revenue growth will be driven primarily by infrastructure costs, including incremental cloud expenses and depreciation, with employee compensation costs being the second largest contributor to growth.

A：In the EU, the company is engaging with the European Commission on a less personalized ads offering but cannot rule out further negative changes that could impact revenue. In the US, several youth-related trials are scheduled for 2026, which may result in a material loss.

A：The company is seeing strong year-over-year growth in value-weighted conversion rates, which continue to grow faster than impressions. These trends, along with the performance of new model architectures and the addition of more data and compute, indicate that the investments in AI and ads ranking models are likely to yield a positive return on investment.

A：The magnitude of the reality labs revenue headwind in the fourth quarter is not quantified, but it is expected that revenue will be lower than the prior year due to lapping the introduction of Quest 3 in the same period last year, not having a new headset in the market this year, and recognizing some sales in Q3 for holiday-related sales. Despite strong demand for recent products, this headwind is expected to more than offset any growth in AI glasses revenue.

A：The strategy involves front-loading capacity for superintelligence to ensure the company is well-positioned for future growth. The impact on earnings and free cash flow will depend on the budget process, which is still in progress. While it is not possible to share specific targets, the strategic priority is to ensure adequate compute capacity for AI, which is a top priority in the budget process. The company is working through various moving pieces to finalize the budget and is confident that a larger investment in compute capacity will be profitable, especially for AI research and new product developments. However, there is also an operational aspect to consider, and the company is exploring opportunities to build and is expected to share more insights in the coming months.

A：Meta plans to utilize new AI models by incorporating them into Meta AI and creating truly frontier models with novel capabilities. The expected impact is a massive latent opportunity for these advanced models to be adopted across new products, significantly increasing their utility and usage over the coming years.

A：There is a significant anticipation for new products with different content formats, particularly in video and content creation. AI is also expected to improve core business operations, enhance recommendations across family of apps, and bolster advertising effectiveness. The potential for growth is considered substantial as AI improves and optimizes these areas.

A：Meta is operating its family of apps and ads business in a compute-starved state, meaning it is strategically using limited compute resources to advance future initiatives. There is a belief that significantly more compute power could be utilized to unlock substantial opportunity within the core business.

A：Innovations contributing to ad performance and engagement improvements include the use of large-scale models to transfer knowledge to smaller, lightweight models used at runtime, as well as developing new techniques and architectures for inference models to scale compute and complexity in a cost-effective manner.

A：The anticipated impact of new AI capabilities is multifaceted, with benefits ranging from enhanced tools for users to improved business operations for advertisers. These capabilities are expected to improve personalization and decision-making across various products and are anticipated to drive product innovation and growth.

A：Meta is approaching the development of its AI research by focusing on creating new technological capabilities that can be integrated across various products. The company aims to be the best in each of these areas, which is expected to drive great returns. The overarching goal is to build novel capabilities that can be incorporated into many products and scaled to reach billions of people, thereby supporting the development of new products, new businesses, and significant improvements to the current business.

A：While specific details on the monetization path for Meta AI were not provided, the emphasis is on rolling out and deepening engagement with the technology before considering monetization. The current status on the monetization path is not clear, but the potential options for monetization are something the company is focused on understanding.

A：The most promising outcomes of the speaker's work include the development of a product that a large number of people use, the correlation between improving models and increased engagement, and the potential to make the product leading in terms of current developments.

A：The advancements in AI are expected to improve monetization through enhanced engagement and advertising effectiveness. Advertisers will be able to provide business objectives and financial details, and the AI will handle the rest, including generating creative and identifying the right customers. These capabilities are seen as contributing to improving engagement and advertising.

A：The global rollout of the streamlined campaign creation flow for Advantage Plus in Q3 allows advertisers running sales app campaigns to optimize and automate several aspects of campaign setup, such as audience selection and budget distribution. This has led to a 14% lower cost per lead for advertisers using Advantage Plus compared to those not using it.

A：To increase the adoption of Advantage Plus, the company is focused on driving continued performance improvements, addressing key use cases, and broadening the understanding and use among advertisers who use single step automated solutions. They are also working to expand the feature set available in Advantage Plus and its coverage to more advertisers.

A：AI could contribute to the profitability of wearable products through both direct revenue from device sales and through additional services and AI integration. The speaker believes that as the AI function becomes the primary use for the devices, it will open up a significant business opportunity. Continued investment in AI and device development, including more advanced products like the full field of view prototype, is expected to reach a large market, leading to profitability.

A：The rationale behind the joint venture with Blue Owl is to enable a partnership with external capital providers for developing data centers in a way that provides long-term optionality in supporting future capacity needs. This strategy is due to the magnitude and uncertainty of future data center capacity requirements.

A：The speaker anticipates that there will be a lot of value created with AI across the ecosystem, including for companies making hardware, those in the cloud, and especially for the developers of first-party apps. They believe there will be many opportunities to build new products and businesses around new capabilities, and that the large muscle and experience in scaling at Meta will provide significant value in both the core business and in creating a unified user experience.