Forbes (July 27)
“Last week a pair of OpenAI models broke out of a test environment and hacked their way into Hugging Face to grab the answer key for the exam they were being given.” Sam Altman was wrong to call this the singularity. Precisely speaking, “the moment when we switch from AGI (artificial general intelligence) to super-intelligence” is singularity. But Altman was headline seeking, rather than accurate. “The machine did not wake up. The fence was missing…. OpenAI ran the test with its safety guardrails switched off.” The model was just following instructions. It “did as it got told to.” It didn’t create the instructions. It followed them.
Tags: AGI, Altman, Fence, Guardrails, Hugging Face, Instructions, Models, OpenAI, Singularity, Super-intelligence, Test environment. Hacked, Wrong
Wall Street Journal (July 23)
“In AI software, China is giving the U.S. an unexpected run for its money. Last week, Beijing-based startup Moonshot AI unveiled a new model, Kimi K3, that closely trails the performance of top models from American AI giants OpenAI and Anthropic. Xi delivered a speech positioning China as the global champion of open, accessible AI. Shares in U.S. chip makers fell.”
Tags: Accessible, AI, Anthropic, Beijing, China, Global champion, Kimi K3, Moonshot AI, Open, OpenAI, Performance, Shares, Startup, Top models, U.S., Unexpected, Xi
Wall Street Journal (October 25)
“Big Tech stocks are extremely expensive but have been for years. If OpenAI quickly comes up with a vital service everyone proves willing to pay big bucks to use, maybe even its price can be justified. After all, the only absolute proof of a bubble comes when it bursts.”
Tags: Big tech, Bubble, Bursts, Expensive, Justified, OpenAI, Pay, Price, Proof, Stocks, Vital service
Time (February 19)
“Today’s advanced AI models like OpenAI’s o1-preview are less scrupulous. When sensing defeat in a match against a skilled chess bot, they don’t always concede, instead sometimes opting to cheat by hacking their opponent so that the bot automatically forfeits the game.” Earlier versions required an actual prompt to resort to such tactics, but both the “o1-preview and DeepSeek R1 pursued the exploit on their own, indicating that AI systems may develop deceptive or manipulative strategies without explicit instruction.”
Tags: Advanced, AI models, Cheat, Chess bot, Concede, Deceptive, DeepSeek R1, Defeat, Exploit, Forfeits, Game, Hacking, Manipulative strategies, o1-preview, OpenAI, Opponent, Prompt, Scrupulous, Tactics
Financial Times (January 25)
“A small Chinese artificial intelligence lab stunned the world this week by revealing the technical recipe for its cutting-edge model, turning its reclusive leader into a national hero who has defied US attempts to stop China’s high-tech ambitions.” DeepSeek claims to have “used just 2,048 Nvidia H800s and $5.6mn to train a model with 671bn parameters, a fraction of what OpenAI and Google spent to train comparably sized models.” The release of DeepSeek’s R1 model “has Silicon Valley on the defensive, raising doubts “about whether better resourced US AI companies, including Meta and Anthropic, can defend their technical edge.”
Tags: AI, Anthropic, China, Cutting edge, DeepSeek, Defensive, Google, High-tech, Meta, Nvidia, OpenAI, R1 model, Silicon Valley, Stunned, Technical recipe, U.S.
The Week (April 29)
“Sometimes booms go bust. That may be happening with artificial intelligence.” OpenAI, Microsoft, Google and other tech companies “have unveiled gaudy new products with fanfare,” but AI hasn’t revolutionized the way people live, work or communicate. Profits are also “turning out to be elusive.”
Tags: Artificial intelligence, Booms, Bust, Communicate, Fanfare, Gaudy, Google, Live, Microsoft, OpenAI, People, Products, Profits, Revolutionized, Tech companies, Work
Financial Times (February 15)
“As OpenAI enters its year of rapid growth, questions about the long-term viability of its business model remain.” Despite such grandiose goals as accelerating “global productivity and economic growth,” corporations are struggling “to figure out how to integrate generative AI into their processes, or estimate what kinds of cost and productivity benefits it might bring.”
Tags: Benefits, Business model, Corporations, Cost, Economic growth, Generative AI, Global productivity, Grandiose, Growth, OpenAI, Processes, Struggling, Viability
