Skip to content

Amazon joins the generative AI race


microsoft, what invested $10 billion in OpenAI, provides access to GPT-3.5, one of the language models powered by ChatGPT, through an application program interface that allows developers to make model access calls directly from their code. Google’s PaLM is equally capable and accessible through its cloud API.

Today, Amazon announced the launch of a platform called Bedrock, which will provide access to cutting-edge language models from Anthropic and AI21two startups developing language models that compete with those of OpenAI and Google.

“It is highly unlikely that there is one model that is the right answer for all customers and for all use cases,” says Selipksy. “We want to offer choice and flexibility.”

Amazon has two generative language models. Titan Text can generate text from a message, and Text Embeddings generate a mathematical representation of text that can be used for tasks such as translation and search.

AWS will also offer access to Stable Diffusion, an AI model for generating images, from AI stabilitya startup that is developing a range of open source generative AI models.

“Bedrock removes significant friction from companies implementing generative models,” says Emad Mostaque, CEO of Stability AI, the company behind the image model. “[It] it is ideal for our open models that can easily access customer data.”

Amazon says Bedrock will allow customers to use its models without the information they input leaking back into the broader pool of data used to train those models, addressing a key concern for organizations that want to build their own AI products. generative. Businesses, particularly those that work with sensitive customer information, such as financial services companies or healthcare providers, are concerned about the security of their data. As users interact with a model, it is sometimes possible for sensitive company information, such as financial reports or business plans, to leak into the model’s training data and be shared with other customers.

“Data leakage in these incredibly powerful and useful AI tools is a major problem,” says Wayne Chang, an entrepreneur working at a stealthy generative AI start-up. “The ramifications are serious.”

Amazon’s generative AI foray will also see it offer new AWS instances optimized for running language models and other generative AI using custom silicon. Amazon is making its artificial intelligence tool for coding, Code Whisperer, free to all individual users through AWS, which can also help. hook users into your cloud.

“Amazon is following their playbook of creating managed service versions of open source tools, but this time also taking closed source models as managed services,” says Nathan Beniach, founder and general partner at Air Street Capital, a AI-focused risk. He adds that while Amazon remains dominant in the cloud, the machine learning tools it offers through AWS don’t appear to have gained significant traction among developers.

Selipsky declined to say how the models measure up to OpenAI’s GPT-4 and Google’s PALM, insisting they are “early days” for the technology in general.



Source link