The multinational eCommerce and digital services firm Amazon has been somewhat conservative in its involvement in AI technology. However, that seems to be changing with their latest announced moves through AWS.
Recently the company launched Amazon Bedrock, a platform to develop AI-powered tools through pre-trained AI generative models; they’ve also rolled out an accelerator program for 10 AI-based startups and have made some of their proprietary AI services available for developers.
Let’s have a closer look.
AWS (Amazon Web Services, the company’s subsidiary for cloud computing and APIs) just debuted Amazon Bedrock –currently on a limited preview. This cloud-based platform hosts proprietary and third-party AI generative models that developers can access via API and use to build new AI generative applications.
The models multilingual text generators Jurassic-2 (from AI21 Labs) and Claude (Anthropic), as well as text-to-image generator Stable Diffusion from Stability AI, are the third-party models available on Bedrock already.
Additionally, customers can access Amazon’s own Titan FM models that include a text generator and an embedding generator (which converts words and phrases into numerical representations containing the semantic meaning of the original text).
Pricing details have not yet been disclosed, although the firm stated in a conversation with TechCrunch that the service is aimed at enterprise customers interested in building large-scale apps.
One of the models hosted on Bedrock is Stability AI’s Stable Diffusion. As you probably heard already, this startup has been sued by the prestigious media licensing firm Getty Images, alleging copyright infringement from Stability AI’s unauthorized use of Getty’s content in the training of their image generative model.
That lawsuit brought to the forefront two debates that experts in copyright and image licensing had sparkled in the early days of visual AI development: whether it’s legal to train AI generative software with other people’s intellectual property without permission or compensation and whether the visuals created with said software could be subject to copyright. Such conversations are ongoing, but so far, it’s looking like the answer to the latter question is no.
Either way, professional users, including brands and businesses, are demanding more legal warranties about AI generative apps before diving in, and Amazon doesn’t seem to be providing entirely satisfactory answers, according to TechCruch. For one, they haven’t disclosed exactly how the Bedrock-hosted models have been trained.
And then, although they made a point to declare their platform has user-focused filters that prevent harmful or illegal content from being created with the models, they also don’t say much about how they protect other people’s copyright or right to privacy. It’s still early, though.
Earlier this month, AWS also announced the creation of a new accelerator program entirely focused on AI generative applications.
This is a ten-week program aimed at AI-based startups from all over the world, and it will provide $300,000 worth of AWS credits to help build their products, high-profile mentorship from leading AI and machine learning experts, industry-focused networking events in the generative AI field, a Demo Day event in San Francisco, and more benefits.
This valuable opportunity for small startups looking to develop new products and services powered by generative AI shows Amazon’s growing interest and involvement in the space.
Amazon is just one of several big names in the global technology industry jumping onto the AI bandwagon, specifically generative AI.
Like Adobe, this firm seems to have been involved in the background for a while but has waited until now that the exponential user and market growth of AI generative tools, thanks to disruptors like OpenAI’s Dall-E and ChatGPT, Midjourney or Stable Diffusion, surpassed everyone’s wildest expectations, to start making bigger and more assertive moves into the field.
Interesting to see how Bedrock will perform. What do you think?
I am an experienced author with expertise in digital communication, stock media, design, and creative tools. I have closely followed and reported on AI developments in this field since its early days. I have gained valuable industry insight through my work with leading digital media professionals since 2014.
AI Secrets is a platform for tech decision-makers to learn about AI technology. Our team includes experts such as Amos Struck (20+ yrs ICT, Stock Photo, AI), Ivanna Attie (expert in digital comms, design, stock media), and more who share their views on AI.
Get AI news in your inbox & join thousands of engineers and managers using AI to boost sales and grow market share