A $30 Billion Funding Round at a $900 Billion-Plus Valuation Is on the Verge of Closing
If you build on Claude or use it in production, the practical takeaway from today's reporting is that Anthropic's compute roadmap is now backed by an enormous fresh war chest. Bloomberg's reporting on the round, co-led by Sequoia, Dragoneer, Greenoaks and Altimeter, would push Anthropic above OpenAI's March valuation for the first time and signals that the company's projected Q2 revenue figure of around $10.9 billion is being treated as credible by investors rather than as a stretch goal.
For users, this matters because it underpins the Microsoft Maia chip rental, the SpaceX compute contract and the continued training of frontier Claude models. In short, it lowers the probability of a near-term capacity squeeze and increases the likelihood of more aggressive feature shipping into Claude Code, the Developer Platform and Cowork over the rest of the year. If you have been hedging your stack on the assumption that Anthropic was the weaker financial bet, that thesis weakened today.
Chris Olah Speaks at the Vatican Alongside Pope Leo XIV for the Release of Magnifica Humanitas
Anthropic posted Chris Olah's remarks today as part of the launch of the first papal encyclical on artificial intelligence, with Olah appearing in person at the Vatican alongside Pope Leo XIV. For most working AI users this is not an immediate product story, but it is worth a moment of attention because it is the clearest public signal yet that Anthropic is positioning Claude as the safety-first option for institutional buyers.
Expect this framing to surface in upcoming enterprise pitches, in the wording of model cards, and possibly in how new safety features are surfaced inside Claude itself. This was a Direct find from anthropic.com/news that had not yet appeared in most mainstream press coverage at time of publication.
Grok V9-Medium, a 1.5 Trillion-Parameter Model, Finishes Training with Public Release in Two to Three Weeks
If you have been waiting to see whether xAI can close the gap with Claude and ChatGPT on day-to-day reasoning and coding tasks, this is the most important xAI update of the quarter. Musk's post confirms that training is complete for the V9-Medium model, that supervised fine-tuning is already underway and that reinforcement learning will start within days, with public release roughly two to three weeks out. The eye-catching detail is that a significant volume of Cursor data was used in supplementary training, strongly suggesting xAI is targeting developer and coding workflows specifically rather than chasing the general assistant market.
V9-Medium triples the parameter count of V8 and is the clearest sign yet that xAI intends to compete on capability rather than just on the X-platform-native experience. For anyone tracking the open-weights versus closed-frontier race, it is also worth watching because xAI's release patterns have in the past included open variants. The post was found on Musk's X account before mainstream press coverage was fully out.
Industry Themes
The dominant pattern in the last 24 hours is financial rather than technical. Anthropic's funding round and the Vatican appearance, both surfaced first on Anthropic's own channels and on Bloomberg before broad press syndication, reinforce a story that has been building all month: the company is consolidating a position as the institutional, safety-oriented option, with the balance sheet to back it.
The second theme is the visible reordering of the xAI roadmap, with Musk's X post on Grok V9-Medium making clear that coding workflows powered by Cursor data are the focus for the next generation, which puts pressure on Anthropic's Claude Code and on OpenAI's Codex agent in particular. The third, quieter, theme is the relative silence from OpenAI on product today, suggesting the next big consumer-facing OpenAI move may be held back until after S-1 paperwork is in.