Anthropic announced on Monday that it would release updated versions of two of its most powerful artificial intelligence (AI) models as the company prepares for a highly anticipated IPO this month.
The new models, Fable 5.2 and Mythos 5.2, can perform up to 30% faster than previous versions, particularly on complex tasks such as coding, while costing 20%, the company said.
Both are built on the same underlying model. Fable 5.2 has safeguards that block most cyber security and biology tasks, which lets Anthropic offer it to all users.
Mythos 5.2 has fewer of those restrictions and is available only to vetted organisations working in cyber security or biomedical research.
The first versions of Fable and Mythos were released under a cloud of controversy. Both models were pulled days after their June launch, when the US Commerce Department, citing national security concerns, ordered Anthropic to bar access by foreign nationals. The controls were lifted weeks later.
The latest versions come as Anthropic is aiming to hold its initial public offering the week of 16 November, when it is expected to raise up to $100 billion, which would make it the largest IPO on record.
The San Francisco-based lab’s Wall Street debut is expected to fetch a valuation exceeding $1 trillion, though estimates have fallen in recent weeks from a high of $2 trillion.
The stock market debut is being seen as a gauge of how long the AI spending boom lifting the US economy can last.
Growth has been staggering at the company. Anthropic’s revenue reached $17 billion in the third quarter, up from just $1.4 billion a year earlier, a securities filing showed earlier this month.
Sales growth slowed from previous quarters, however, amid fierce competition from ChatGPT-maker OpenAI and cheaper models, many of them from China.
Anthropic also posted an operating loss of $567 million in the third quarter as it paid more for short-term computing capacity.
The company’s valuation has also been weighed down by a broader market pullback after a string of quarterly earnings reports spooked investors.
These raised concerned about the hundreds of billions of dollars that US tech giants are spending on AI. Democrats regained control of Congress in the US midterms last week, raising investor concerns about a potential regulatory crackdown on the tech industry, including against data centre buildouts.
Ballot measures restricting data centre construction could hamper the growth plans of AI developers such as Anthropic, as well as rivals including OpenAI, Meta, Microsoft and Google.
AFP
Ltd
