US Edition‘AI Race Is Commoditising’: Oracle’s Cofounder Bets Big on Infrastructure and Enterprise Data
3 Articles
3 Articles
‘AI race is commoditising’: Oracle’s Cofounder bets big on infrastructure and enterprise data
Oracle cofounder Larry Ellison says the AI model race is already commoditising, with rivals training on the same public data. Oracle is betting $50 billion on AI infrastructure and enterprise data, even as competition from cloud giants intensifies.
What is one of the problems of the current AI? Models, mainly LLMs, use roughly the same data and, de facto, all the big LLMs in the market are little or no identical.They use the same data to be trained, with variants here and there. Larry Ellison said that a few days ago.There would therefore not be enough variation between the LLMs of Google, Meta, Open AI, etc. "In order for these models to reach their full potential, it is necessary to trai…
Larry Ellison, co-founder and technical director of Oracle, has pointed out what he considers to be the fatal flaw of the current AI race: all major models, from ChatGPT to Gemini and Llama de Meta, are trained from the same public data available on the Internet. So they are all basically identical. That's why they're becoming commonplace so quickly. Ellison's solution? The next gold rush will not be to build better models.
Coverage Details
Bias Distribution
- 100% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
