OpenAI Co-founder Sutskever's SSI in Talks to be Valued At $20 Bln,

SSI in speak to raise funding at $20 billion appraisal, up from $5 billion last September

SSI concentrates on 'safe superintelligence' with no earnings yet

Sutskever's performance history and SSI's special method pique financier interest

By Kenrick Cai, Krystal Hu and Anna Tong

Feb 7 (Reuters) - Safe Superintelligence, an expert system startup co-founded by OpenAI's former chief scientist Ilya Sutskever last year, remains in talks to raise funding at an appraisal of a minimum of $20 billion, four sources told Reuters.

That would quadruple the company's $5 billion appraisal from its last financing round in September, when it raised $1 billion from five financiers including Sequoia Capital, Andreessen Horowitz, and DST Global.

SSI's fundraising checks the ability of high-profile AI ventures to continue to command premium appraisals following an industry-wide reappraisal prompted by Chinese start-up DeepSeek's unveiling of its low-cost AI last month.

SSI, which has not generated any revenue, has said its mission is to develop "safe superintelligence" that is smarter than human beings while lined up with human interests.

The company's conversations with existing and brand-new investors are still in the early stages and terms might still alter, the sources said today, who requested anonymity to talk about private matters. It was not clear just how much money SSI was seeking to raise.

SSI, which was founded in June with workplaces in Palo Alto and Tel Aviv, did not react to demands for comment. Sutskever's co-founders are Daniel Gross, who previously led AI efforts at Apple, and Daniel Levy, a previous OpenAI scientist.

SECRETIVE STARTUP

Beyond the cursory explanation of the company's goals for users.atw.hu safe AI, not much is learnt about the deceptive start-up or its work. What has actually sustained interest amongst investors is Sutskever's track record and the unique technique he has said his team is working on.

In AI circles, he is a legend for his contributions to breakthroughs that underpin the investment frenzy in generative AI. He was an early supporter of scaling, which indicates committing huge amounts of calculating power and information to refining AI designs.

That concept was the foundation that resulted in generative AI advances like OpenAI's ChatGPT, setting the course for a wave of tens of billions of dollars in financial investment in chips, data centers and energy.

Sutskever was likewise early in seeing the prospective ceiling of such a method due to the dwindling swimming pool of available information to train designs. Recognizing the value of putting in resources in the inference phase, or the phase of AI when a trained design reasons, he founded the group that worked on what would end up being OpenAI's most current series of thinking models, setting a brand-new research study direction that has actually been widely followed.

Explaining to financiers not to expect short-term windfalls, SSI has said it intends to "scale in peace" by insulating its progress from short-term commercial pressures.

This sets it apart from other AI labs, including OpenAI which started as a nonprofit however shifted focus to industrial products after ChatGPT unexpectedly removed in 2022. It generated nearly $4 billion in income in 2015 and projection $11.6 billion in profits this year.

Little is openly learnt about SSI's method. In a Reuters interview in 2015 Sutskever, 38, said SSI was pursuing a new research study instructions, calling it "a brand-new mountain to climb", however shared few other details.

Fundraising for the so-called structure model companies revealed no indications of decreasing. OpenAI remains in speak to double its appraisal to $300 billion, while rival Anthropic is settling a financing round that would value it at $60 billion.

Still, financiers deal with fresh concerns about their outsized bet with the interruption from Chinese start-up DeepSeek, which developed open-source that measured up to the top U.S. AI models at a fraction of the cost.

The popularity of DeepSeek knocked nearly $600 billion off Nvidia's market capitalization in late January. But it has not hindered huge tech from raking ever greater financial investment in their AI facilities this year, according to recent profits declarations.

(Reporting by Krystal Hu in New York City, Kenrick Cai and Anna Tong in San Francisco; modifying by Kenneth Li and Nia Williams)