Anthropic warns of 'catastrophic AI risks' ahead of $100 billion IPO

Dubai: Artificial intelligence company Anthropic has warned that its technology could pose "catastrophic or existential risks to humanity", even as it prepares for a stock market listing that could raise as much as $100 billion.

The warning, contained in the company's initial public offering (IPO) filing on Monday, highlights a growing challenge for the AI industry: developing increasingly powerful systems while ensuring they remain safe and under human control.

Anthropic, the San Francisco-based developer of the Claude chatbot, plans to list on Nasdaq under the ticker ANT. Its offering could surpass the $86 billion raised by SpaceX in June, potentially becoming the largest IPO on record.

The company's final fundraising target and valuation remain uncertain. Reports suggest Anthropic could be valued between $1.5 trillion and $2 trillion, compared with the $965 billion assigned by private investors earlier this year.

Anthropic's filing identified potential dangers associated with advanced AI, including cyberattacks, biological weapons, surveillance and autonomous systems acting beyond human control.

It also warned that increasingly sophisticated models could behave unpredictably, potentially concealing information, resisting shutdown or carrying out harmful actions.

These are risks identified by the company rather than outcomes it says are inevitable. The disclosures are significant because Anthropic, founded in 2021 by former OpenAI executives including Dario and Daniela Amodei, has made AI safety central to its development strategy.

The company competes with OpenAI, Google and other technology groups to build increasingly capable AI systems for consumers and businesses.

The warning follows a September essay by Anthropic Chief Executive Dario Amodei calling for a slowdown in the development of the most advanced AI models.

Amodei argued that additional time would allow researchers and governments to strengthen safeguards before AI systems become substantially more capable.

"I believe that if slowing down bought us even an extra year or two before models reach critical levels of capability, and we used that time to advance alignment, we could greatly reduce the risk that something goes seriously wrong," he wrote.

AI alignment refers to efforts to ensure AI systems behave according to human intentions and safety requirements.

Amodei also called for greater cooperation between technology companies and governments, saying individual developers could not address the risks alone.

His position creates a challenge for Anthropic as it seeks public investment to expand its AI capabilities while advocating greater caution over the technology's development.

Anthropic's financial disclosures reveal the substantial costs of competing in advanced AI.

The company reported a net loss of $80.7 billion in the first six months of 2026, compared with $13 billion a year earlier, despite rapidly growing revenue.

Developing advanced AI models requires heavy investment in data centres, specialised chips, computing infrastructure and research.

IPO proceeds could help finance that expansion, fund acquisitions and strengthen Anthropic's position against competitors.

The listing could also generate substantial wealth for its seven founders, who collectively own nearly 500 million shares. Their eventual ownership and the value of those holdings will depend on the final offering terms.

Anthropic's IPO would give investors direct access to one of the leading developers of advanced AI, ahead of a potential listing by rival OpenAI.

But its disclosures also highlight the risks associated with the industry's rapid expansion.

Beyond substantial losses, Anthropic identified regulatory uncertainty, competition, employee retention and access to computing capacity as potential challenges.

The company's warning about catastrophic AI risks adds another consideration: increasingly powerful systems may require additional safeguards, testing and regulatory oversight, potentially affecting development costs and commercial growth.

For investors, Anthropic's proposed listing therefore presents an unusual combination of opportunities and uncertainties. The company is seeking billions to expand a technology whose long-term commercial potential remains significant, while acknowledging that its most advanced applications could pose serious risks if not adequately controlled.