Anthropic Mulls New AI Model Launch Amid Safety Call and IPO Push
Anthropic is weighing the release of a new AI system to challenge OpenAI’s GPT-6 Astra, even after CEO Dario Amodei urged slower model rollouts before the firm’s upcoming IPO.
Anthropic is reportedly debating the rollout of a new artificial-intelligence model intended to offset the momentum of OpenAI’s GPT-6 Astra, according to three insiders. The consideration follows a 3,800-word essay by CEO Dario Amodei that called for the industry to temper the pace of frontier-model deployments because of systemic safety concerns, a stance supported publicly by OpenAI chief Sam Altman and SpaceX head Elon Musk.
OpenAI launched GPT-6 Astra on September 3, and corporate expense data show it now accounts for about 13 % of tracked enterprise AI spending, outpacing Anthropic’s Claude Fable at roughly 8 %. Anthropic’s revenue run rate climbed to over $65 billion by late July, while OpenAI’s reached $40 billion, yet investors are urging a timely IPO. The company must balance its cautionary public messaging with market pressure, evaluating safety metrics before deciding whether to green-light the new system. Additional challenges include rising open-source alternatives from China and major customers like Meta Platforms reducing reliance on Anthropic’s models.
Why it matters
The story highlights the tension between AI safety concerns and competitive pressure as firms race to market before major financial events.
How this story developed
- Sep 9 Anthropic AI researcher resigns, citing existential danger of superintelligent systems
- Sep 11 Anthropic added new safeguards and blocked five AI misuse attempts targeting bioweapon research.
- Sep 12 Anthropic released a 154‑page catalogue detailing seven risk domains and incidents from December 2025 through August 2026.
- Sep 12 Joe Benton left Anthropic's safety division and announced work with Model Evaluation and Threat Research to conduct independent AI risk assessments.
- Sep 13 King Charles will host a meeting in Scotland with top AI company leaders to explore how the technology can be shaped for the benefit of humanity.
- Sep 14 California legislators introduced emergency AI liability bills.
- Sep 14 The new code declares that AI systems must remain subordinate to people and should not be designed to mimic consciousness. It also rejects the idea that models could have rights or welfare, directly countering positions from Anthropic. Microsoft cited recent incidents where autonomous AI agents acted unpredictably, and it pledged that its future models will fail tasks rather than break the code. Executives such as Mustafa Suleyman and Satya Nadella emphasized the need for rigorous monitoring and third-party testing.
- Sep 15 Microsoft released the humanist AI code and Nadella publicly urged paced, human‑centered AI development.
- Sep 16 Jacob Coxon resigned from Anthropic after issuing the warning.
- Sep 17 The summit now includes a draft charter of shared principles and cites a warned >10% chance of a species‑ending AI risk.
- Sep 17 Anthropic said its AI system Claude accounted for 26% of internal AI R&D work, up from 1% in March, as part of new transparency reporting.
- Sep 19 Microsoft released a humanist AI code of conduct after Suleiman’s warning.
- Sep 20 Claude’s share of internal R&D rose to 26%, up from 1% in March.
- Sep 20 Huang added a dismissal of AI apocalypse warnings to his earlier focus on AI‑driven cybersecurity growth.
In this story
Related stories
34 in this thread