The 400-page prospectus devotes over 90 pages to dangers particular to an trade that has confronted a string of studies in current months of AI techniques behaving in surprising and dangerous methods.
It additionally displays the dominant perception amongst researchers inside Anthropic that AI must be developed by a secure pair of palms, or in any other case see its extraordinary powers misused with China having the sting.
The submitting cited considerations that superior techniques “may pose catastrophic or existential dangers to society if not developed and deployed responsibly”.
Anthropic additionally warned that its “engagement with policymakers and public positions on AI regulation and security may topic us to heightened scrutiny, political backlash, or disproportionate regulatory motion”.
The corporate has been focused in current months by the administration of US President Donald Trump for refusing to carry restrictions on how the Pentagon may use its know-how, and the Protection Division has since labelled it a nationwide safety threat and barred it as a provider.
Regardless of these challenges, Anthropic’s enterprise has grown in a short time. Gross sales reached US$11.6 billion within the April-June quarter, up from US$787 million a 12 months earlier. That compares with US$4.6 billion in gross sales for all of 2025.
The corporate made its first working revenue in that quarter, near US$1 billion. It nonetheless reported an total lack of US$80 billion for the primary half of the 12 months, principally due to accounting expenses linked to its rising worth somewhat than money spent.
A lot of the expansion comes from companies utilizing Claude to put in writing software program and deal with workplace work.
Practically 12,000 organisations now spend a minimum of US$100,000 a 12 months with Anthropic, up from fewer than 1,000 a 12 months in the past.
