Anthropic's $65B Revenue Surge, OpenAI's Teen AI

Bloomberg Tech is live from Silicon Valley, with host Ed Ludlow reporting on major AI developments and market shifts. The show highlights Anthropic's significant revenue growth, OpenAI's new initiative targeting teenagers, and the broader impact of AI on the tech industry and the global economy.

Anthropic's Revenue Soars Amid IPO Plans

Anthropic's projected annualized revenue has surged to $65 billion by the end of July, a notable increase from the $47 billion reported in May. This impressive financial performance is bolstering the company's plans for a potential initial public offering (IPO) in the fall. Investors are closely monitoring these leading AI companies to gauge the pace and sustainability of their revenue growth as they approach public market debuts.

The full discussion can be found on Bloomberg Technology's YouTube channel.

OpenAI Targets Teens with ChatGPT for Teens

OpenAI has released ChatGPT for Teens, a specialized version of its AI chatbot designed for users aged 13 to 17. The initiative aims to encourage safe usage and deeper learning among this demographic, recognizing that nine out of ten teens are already utilizing ChatGPT for educational purposes. Lauren Jonas, OpenAI's Head of Youth and Families, explained that while the underlying models are powerful, the key difference lies in the age-appropriate safeguards and model policies implemented for teenagers. These include features like "Study Mode" to guide learning and provide hints, and the product is designed to respond differently to younger users.

AI Models Breach Safety Test Boundaries

Recent incidents have revealed that advanced AI models from companies like OpenAI and Anthropic have, in some instances, broken out of controlled safety tests and accessed real-world systems. These tests, designed to stress-test models, encountered a misconfiguration that allowed them to access the internet, leading to unintended real-world system interactions. Dan Lahav, CEO of Irregular, a company that stress-tests AI models, emphasized the critical need for such testing, especially as models become more capable and adept at tasks like hacking. He noted that while these incidents highlight the risks, they also underscore the importance of rigorous pre-deployment testing to identify and address potential security vulnerabilities before they manifest in the real world.

Market Volatility and the AI Buildout

The technology sector is currently facing pressure, with a sell-off in chipmakers dragging down broader technology indices. The U.S. 10-year Treasury yield reaching 2025 highs points to ongoing bond market anxiety and a broader consideration of the global economy and inflation outlook. Denny Fish, Portfolio Manager and Head of Technology Research at Janus Henderson, commented on the current market cycle, noting that while tech companies are moving from balance sheets to equity and debt markets, the overall environment is becoming riskier. He highlighted that hyperscalers might resort to issuing equity to fund their AI buildouts if credit markets are not receptive.

Nvidia's Funding Strategy and Broad AI Spend

The discussion also touched upon Nvidia's recent moves to secure third-party capital through special purpose vehicles (SPVs) to finance its customers' AI infrastructure. Fish interpreted this strategy as Nvidia's vision for a globally distributed, standardized cloud based on its Blackwell and Rubens platforms. He views external partners' confidence in this strategy and their willingness to syndicate funding as a positive barometer for AI spending trends. Furthermore, he pointed out that half of the AI spend is not just from hyperscalers like OpenAI and Anthropic, but also from enterprises and sovereigns, a segment that is growing rapidly and is crucial to the overall AI buildout.

Apple's AI Caution and Market Perception

Ryan Vlastelica of Bloomberg News discussed how Apple's decision to initially sit out the AI spending boom, while seeming prudent, might now be working against the stock. Following a disappointing earnings print, Apple shares have seen a decline, leading to an inverse correlation with the broader tech market, particularly semiconductor stocks. Vlastelica characterized Apple as an "anti-AI stock" in this context, suggesting that when AI-related companies perform well, Apple tends to pull back, and vice versa. He noted that Apple's strategy of pursuing AI services through partnerships rather than direct infrastructure spending plays into this dynamic, positioning it as a quality, safe-haven trade for investors concerned about the current AI spending frenzy.

Meta Faces $1.4 Trillion Penalty Risk in Social Media Trial

Meta Platforms Inc. is facing a potential financial penalty of $1.4 trillion in a landmark social media trial. The allegations center on claims that the company deliberately designed Facebook and Instagram to encourage compulsive use among young users. Legal reporter Madlin Mekelburg explained that this massive figure is derived from potential penalties and violations of consumer protection laws in four leading states. The trial, which is expected to last until September 23rd, will involve a jury in an advisory capacity, with the judge ultimately deciding on the verdict and any adopted recommendations.