OpenAI has acquired the domain name Chat.com for an undisclosed amount, TechCrunch reports. The domain, which was registered in 1996 and is one of the oldest on the web, now redirects to OpenAI’s AI chatbot, ChatGPT. According to a post on X-Twitter by HubSpot co-founder and CTO Dharmesh Shah, who sold the domain last year …
Nvidia has become the world’s most valuable company, surpassing Apple with a market cap of $3.43 trillion as of November 5, 2024. According to Ryan Vlastelica reporting for Bloomberg, Nvidia’s dominance reflects the immense impact of artificial intelligence on Wall Street. The chipmaker is responsible for a quarter of the S&P 500’s 21% gain this …
Google subsidiary DeepMind has introduced SynthID-Text, a system for watermarking text generated by large language models (LLMs). By subtly altering word probabilities during text generation, SynthID-Text embeds a detectable statistical signature without degrading the quality, accuracy, or speed of the output, as described by Pushmeet Kohli and colleagues in the journal Nature. While not foolproof, …
Anthropic has released its newest and smallest AI model, Claude 3.5 Haiku, which outperforms the previous flagship model, Claude 3 Opus, on various benchmarks at a lower cost, according to the company. However, Anthropic has increased the pricing for Claude 3.5 Haiku to reflect its enhanced capabilities, with input tokens now costing $1 per million …
Researchers from UC San Diego and Tsinghua University have developed a method that improves artificial intelligence’s ability to understand when to use external tools versus relying on built-in knowledge, similar to how human experts approach problem-solving. Using a relatively small language model with 8 billion parameters, the team achieved a 28% improvement in answer accuracy …
OpenAI, the $157 billion artificial intelligence company founded in 2015 as a nonprofit, is in preliminary discussions with the California and Delaware attorneys general to convert to a for-profit public benefit corporation. According to two people familiar with the matter, the process will likely involve regulators examining how OpenAI values its intellectual property assets, such …
Meta is making its Llama AI models available to U.S. government agencies and contractors working on defense and national security applications. According to a blog post by Meta cited by TechCrunch, the company is partnering with firms like Accenture, Amazon Web Services, and Lockheed Martin to bring Llama to these entities. The move comes after …
OpenAI CEO Sam Altman announced during a Reddit AMA on Thursday that the company likely won’t release GPT-5, its next major AI model, in 2023 due to the increasing complexity of these models. Altman cited “limitations and hard decisions” in allocating compute resources as the reason for the delay, despite “some very good releases coming …
Generative AI is blurring the distinction between content and data, argues Robert Rose in an article for the Content Marketing Institute. As AI gains prevalence, companies will need people in new roles to guide how AI-generated insights are transformed into meaningful value rather than settling for “good enough” AI-produced content. Rose asserts that while AI …
In an experiment, journalist Kashmir Hill let AI chatbots and tools make daily decisions for her for one week, as described in The New York Times. The generative AI systems saved time by quickly offering advice on meals, clothing, activities, and more, but their choices seemed to favor blandness and statistical averages over individuality. While …