OpenAI mulls product changes
OpenAI co-founder Sutskever's SSI valued at $20B
Nvidia not rattled by DeepSeek
You’re Doing Breakfast Wrong
Discover Huel Black Edition—complete nutrition without compromise.
With 40g of protein and 27 essential vitamins, it’s the perfect meal for your busy life. Whether at home or on the go, fuel up in seconds with a high-protein meal.
Even better? New customers get a bundle of savings to kickstart their Huel journey. Use code BEHUEL15 for 15% off your first order, plus a FREE t-shirt and shaker.
***
OpenAI mulls product changes
In recent public statements, OpenAI's top executives announced plans to unveil more of the detailed steps their models take to reason through complex queries, akin to DeepSeek's R1 model. OpenAI is also re-evaluating its stance on open source after years of closely guarding the code behind its products. CEO Sam Altman described the competition as "invigorating" and pledged to release more impressive models soon to demonstrate OpenAI's continued leadership in the AI race. The company has already made strides by launching an updated version of its o3 reasoning system last week. With the o3-Mini version available to free subscribers by selecting Reason in app.
This shift in OpenAI's rhetoric underscores the early influence of DeepSeek on the global AI landscape. Within Silicon Valley, skepticism persists regarding DeepSeek's claims of developing its model at a fraction of the cost others incur. However, there are evident concerns about DeepSeek and China's overall progress in the AI field. Several companies have recognized the need to learn from DeepSeek's viral success, including its chatbot design choices and innovative approaches to managing with fewer high-end computer chips. Meta CEO Mark Zuckerberg recently acknowledged on an earnings call that DeepSeek's novel advancements are being closely examined and potentially integrated into their systems, highlighting the dynamic nature of competition in the AI sector.
Nowhere is this shift more apparent than at OpenAI, particularly concerning open source. By adopting a more open approach, DeepSeek has the potential to expand its chatbot's reach in Western markets. By facilitating easier adaptation of the underlying technology for other developers. Despite its name, OpenAI has historically preferred closed or proprietary AI models for competitive and safety reasons. But in an “Ask Me Anything” discussion on Reddit last week, Altman said: “I personally think we have been on the wrong side of history here, and need to figure out a different open-source strategy.” To that end, he said, OpenAI is now discussing whether to release its model weights, a key part of AI systems like GPT....
***
The gold standard of business news
Morning Brew is transforming the way working professionals consume business news.
They skip the jargon and lengthy stories, and instead serve up the news impacting your life and career with a hint of wit and humor. This way, you’ll actually enjoy reading the news—and the information sticks.
Best part? Morning Brew’s newsletter is completely free. Sign up in just 10 seconds and if you realize that you prefer long, dense, and boring business news—you can always go back to it.
***
OpenAI co-founder Sutskever's SSI valued at $20B
Safe Superintelligence, an artificial intelligence startup co-founded by former OpenAI chief scientist Ilya Sutskever last year, is currently negotiating funding at a valuation of at least $20 billion, according to four sources who spoke to Reuters. This would represent a fourfold increase from the company's previous valuation of $5 billion during its last funding round in September, when it secured $1 billion from five investors, including Sequoia Capital, Andreessen Horowitz, and DST Global. SSI's efforts to raise funds come amid an industry-wide reassessment triggered by Chinese startup DeepSeek's launch of a low-cost AI system last month. Despite not yet generating any revenue, SSI remains committed to its mission of developing "safe superintelligence" that surpasses human intelligence while remaining aligned with human interests.
The startup remains largely enigmatic, with limited information available about its operations. Investor interest is primarily driven by Sutskever's stellar reputation and the innovative approach he claims his team is pursuing. In AI circles, Sutskever is revered for his pivotal contributions to generative AI advancements, which have spurred a surge in investments in AI-related technology. He was an early proponent of scaling, dedicating vast amounts of computing power and data to refine AI models, a foundation that led to innovations like OpenAI's ChatGPT.
Sutskever also foresaw the limitations of this approach due to the dwindling availability of data to train models. Emphasizing the importance of resources in the inference stage—when a trained model makes predictions—he led the team that developed OpenAI's latest series of reasoning models, setting a widely adopted research direction. Fundraising for foundation model companies shows no signs of stopping. OpenAI is in discussions to double its valuation to $300 billion, while competitor Anthropic is finalizing a funding round that would value it at $60 billion….
***
Nvidia not rattled by DeepSeek
Nvidia's long-term strategy hinges on the belief that China will remain a crucial market for years to come. As the world's leading manufacturing powerhouse, China is poised to be at the forefront of AI-related fields such as robotics and autonomous driving. According to Nvidia insiders, maintaining business ties in China is essential for the company to stay relevant, especially as AI functions become increasingly integrated into everyday devices manufactured in the country.
Chinese headhunter Xie Weide revealed that some companies have been offering Nvidia engineers and marketing managers double their current salaries. For months, Xie frequented a coffee shop near Nvidia's Shanghai office during lunchtime, approaching staff to recruit them for Huawei and another Chinese company. At a recent gathering in Beijing, Nvidia CEO Jensen Huang proudly stated that Nvidia's annual employee turnover in China was just 0.9%, lower than the company's global average of 2%. "Once you join Nvidia, you don't leave," Huang said. "If you join Nvidia, you're going to grow old with me."
The conflict between Nvidia and U.S. officials over its China business began in earnest in late 2022, coinciding with the AI frenzy sparked by OpenAI's chatbot ChatGPT. Nvidia's chips were in high demand among top AI players like OpenAI, Google, Amazon, Microsoft, and Meta. In October 2022, U.S. put in place the first export controls specifically targeting AI chips, including the A100s DeepSeek had bought—setting off a cat-and-mouse game of regulations to limit the spread of the chips. Nvidia’s engineers soon rolled out a chip called A800—a variant of the A100 that met U.S. rules and that Nvidia would effectively sell only within China. The development came at breakneck speed for an industry where new chips often take years….
Cheers! SBalley Team