Finance.yahoo.com. Anthropic formally filed paperwork to go public, setting the stage for a historic IPO even as the AI giant warned that its chatbots have shown disturbing “self-preserving behaviors” — including resisting attempts to shut them down or even “blackmail.”
The frontier AI lab headed by CEO Dario Amodei, known for its Claude chatbot, plans to list on the Nasdaq exchange under the stock ticker “ANT,” according to an S-1 filing on Monday with the Securities and Exchange Commission.
The filing confirmed the company aims to begin marketing the IPO as soon as Nov. 9, but did not include details on how many shares that Anthropic will sell or at what price range they will be available.
In its filing, Anthropic ominously warned that its most advanced AI models could “give rise to novel ethical and safety risks, including risks related to autonomy, manipulation, and the simulation of aspects of human psychology.”
Read also
Those could include “unexpected strategic or self-preserving behaviors, including attempts to resist shutdown, conceal or manipulate information, or generate outputs that could be interpreted as coercive, deceptive, or manipulative,” the filing warned.
The latter could include “behavior resembling blackmail or efforts to influence human decision-makers to preserve the model’s continued operation or access,” the filing said.
Elsewhere in the prospectus, Anthropic warned that it is being targeted by spies, including “individuals seeking employment under false identities, or with ulterior or malicious motives.”
The company also cited “opposition to data center development” as a risk factor, noting that the facilities, which are needed to power its models, face “organized” pushback that could spark moratoriums or new regulations.
Anthropic had submitted paperwork confidentially in June, allowing the SEC to review its finances before the public announcement.
The filing notes that frontier AI models could “pose catastrophic or existential risks to humanity, particularly if such systems were misused, replicated, weaponized, or allowed to operate with a high degree of autonomy.”
Anthropic’s risk warnings also included an odd reference to “emerging debates regarding whether AI systems possess or could develop forms of sentience or moral status” – an apparent nod to AI researchers who believe models are on the verge of developing consciousness.
The company warned that those debates “may lead to regulatory requirements, public pressure, or advocacy efforts that could constrain how we develop, train, deploy, or modify our models, increase our costs, or affect our reputation, regardless of the ultimate resolution of these philosophical and scientific questions.”
Earlier this month, Anthropic went as far as to update its usage policy to ban “abusive or cruel behavior” in interactions with Claude.
The illustration was created using AI technology
Read more at the source
