Anthropic has unveiled Claude 3, a suite of AI models that are faster and more powerful than previous versions. Claude 3 Opus outperformed OpenAI’s GPT-4 and Google’s Gemini Ultra on benchmark tests, and the new models offer multimodal support for analyzing unstructured data. Anthropic, backed by Google, Salesforce, and Amazon, has become a leading AI startup, securing $7.3 billion in funding. The generative AI field has grown significantly, with a record $29.1 billion invested in 2023. Claude 3 can summarize up to 150,000 words and has a better understanding of risk in responses.
Anthropic’s Claude 3 AI models will be available on Amazon Bedrock. This expansion of the collaboration between AWS and Anthropic will see AWS become the primary cloud provider for safety research and model development. Amazon Bedrock already has a strong user base, and this addition further strengthens its generative AI offerings.
More than 100 top AI researchers have signed an open letter urging generative AI companies like OpenAI, Meta, and Midjourney to allow independent researchers access to their systems. The researchers argue that strict protocols and opaque company rules hinder safety testing and independent research. They call for legal and technical safe harbors for researchers and warn against repeating social media platforms’ mistakes in banning certain types of research.
With all the talk of OpenAI, I don’t want to miss moves by Anthropic. They are a significant player in the space, and our role as service providers is to make recommendations on which models and companies are the ones to focus on.
So we’re keeping up with their moves.