Matt Wood, VP of AI, announced his departure from AWS after 15 years amid the company's potential lack of progress in the generative AI boom. AWS reportedly passed on opportunities to back leading AI startups like Cohere and Anthropic.
Garman sees price as an AWS advantage, given its projects to develop custom silicon for running and training models. Investors are, however, increasingly skeptical that generative AI trials by Big Tech are paying off.
In a recent Gartner poll, 49% of companies said that demonstrating value is their top barrier to generative AI adoption while Gartner predicts that a third of generative AI projects will be abandoned after the proof of concept phase by 2026.
Anthropic CEO Dario Amodei has written a lengthy analysis claiming AI risks can be mitigated, resulting in social uplift and renewed prosperity.
A team of AI security experts from the UK's AI Safety Institute and Grey Swan AI have produced a dataset called AgentHarm to determine the harmfulness of popular AI models such as OpenAI's GPT-4o.
Anthropic has updated its Responsible Scaling Policy (RSP), with two groups of models mentioned needing upgraded safeguards before deploying. Anthropic is hiring a head of responsible scaling.
Adobe has launched video generating capabilities for Firefly AI and has also introduced Project Super Sonic, which uses AI to generate sound for footage.
Amazon's new Sequoia automated storage and retrieval system has been introduced by its robotics arm.
Chinese researchers from Kuaishou Technology, Peking University and the Beijing University of Posts and Telecommunications have come up with a video generator model called Pyramid Flow.
Matt Garman of AWS is acquiring AI start-ups and investing in training systems like Olympus to boost progress in generative AI, which is said to be a crucial aspect for AWS moving forward. AWS is currently in a good position thanks to a series of custom silicon, including 'Trainium' chips.