Buyer beware: OpenAI’s o1 large language model is an entirely different beast
TL;DR: OpenAI’s new o1 model marks a significant leap in AI reasoning capabilities but introduces critical risks. Its ...
DeepSeek open-sources new AI model with 671B parameters
Chinese artificial intelligence developer DeepSeek today open-sourced DeepSeek-V3, a new large language model with 671 ...
Chinese AI model maker Stepfun raises hundreds of millions in Series B funding
Stepfun is just the latest Chinese model maker to rake in millions of dollars in funding. One week ago, a startup called ...
unite7d
How Large Language Models Are Unveiling the Mystery of ‘Blackbox’ AI
AI is becoming a more significant part of our lives every day. But as powerful as it is, many AI systems still work like ...
Why AI language models choke on too much text
Large language models represent text using tokens, each of which is a few characters. Short words are represented by a single ...
How OpenAI’s Jerry Tworek found a new way forward for large language models
The veteran AI researcher was a key developer of OpenAI’s o1 model, which 'thinks' longer and harder about problems.
Microsoft’s Phi-4 (14B) AI Model Tested Locally: Performance, Limitations and Future Potential
Explore Phi-4's potential and limitations in AI tasks like structured data handling, code generation, and ambiguous question ...
ReadWrite8d
AI model displays alignment faking, new Anthropic study finds
The researchers suggest the demonstration of sophisticated alignment faking in a large language model “should be seen as a ...
COBIT8d
Navigating the Complex Landscape of Large Language Model Security
A multi-faceted approach, including robust model training, is needed to effectively deal with large language model ...