OpenAI closed its largest funding round on record at $6.6 billion, valuing the company at $157 billion—a tenfold increase since 2021. The round was led by Thrive Capital, with participation from Microsoft, Nvidia, SoftBank, and others. What matters more than the headline number: these funds flow directly into compute infrastructure and model training, not marketing. OpenAI projects $3.6 billion in annualized revenue by year-end, scaling to $11.6 billion by 2025, but faces $5 billion in annual operating costs. The company is still unprofitable, yet investors price in a pathway to margins if revenue growth holds and costs stabilize. A structural wrinkle: OpenAI must convert from its hybrid nonprofit-for-profit structure to pure for-profit status by 2027, a condition tied to this round. The shift prompted departures from senior research staff, including CTO Mira Muratti's brief absence during Sam Altman's contested ouster last year. What this signals is not hype but mathematical confidence—a $157 billion valuation on a 43x revenue multiple is aggressive but defensible for a scaling AI infrastructure company.
Anthropic, OpenAI's closest competitor, sits at a $40 billion valuation with projected $1 billion revenue, trading at a 40x multiple. This parity in pricing suggests the market has settled on two dominant players in large language models, each with differentiated positioning: OpenAI leads on capability and adoption (250 million weekly ChatGPT users); Anthropic competes on transparency, ethics, and developer-focused tooling. Both will likely remain because both generate real revenue. The broader AI race—Google embedding models into hardware, Meta open-sourcing Llama, smaller startups attacking niche problems—creates genuine competition that accelerates improvement. Hype there will be; waste too. But if you have a defensible angle into this market, capital exists to back it.
Geoffrey Hinton and John Hopfield won the 2024 Nobel Prize in Physics for foundational work on neural networks and machine learning—a recognition that AI no longer belongs to computer science alone. Hinton, the "godfather of AI," led the 2012 AlexNet breakthrough that proved convolutional networks could outperform statistical methods on image recognition. That moment, replicated across domains, opened the floodgates for current AI investment. Hopfield's work on associative memory in neural systems provided parallel theoretical grounding. The Nobel committee's choice reflects something less discussed in tech circles: AI's relevance to physics itself, where neural networks have accelerated materials research and simulations. This award legitimizes the entire enterprise at a moment when critics question AI's utility.
On obesity, new data suggests the US may have passed peak weight. Obesity rates fell from 40.2% to 40% in 2023—the first documented decline in decades. Credit goes to GLP-1 receptor agonists like Ozempic and Wegovy, drugs originally developed for Type 2 diabetes that produce sustained weight loss. One in eight American adults has used one for weight loss; current users sit at 6% of the population. Educated, affluent cohorts show steeper decline, indicating both awareness and access disparity. The parallel to tobacco is instructive: cigarette consumption peaked in 1963, followed decades later by declining lung cancer rates once policies (taxation, restrictions, public health campaigns) took hold. If obesity decline sticks, cardiovascular disease and diabetes mortality should follow a similar lag—potentially adding years to average lifespan globally. The caveat is stark: these drugs are expensive, unavailable in most of the world, and do not address the underlying driver—food environment and eating behavior. A pharmacological solution to a systemic problem will help millions but won't solve the problem.