in ,

Amazon introduces tools for scaling AI apps, enhancing accuracy. #GenerativeAI

Amazon adds tools for scaling generative AI applications — and improving accuracy issues

Amazon announced new tools at the AWS Summit New York to make generative AI applications easier to create and more accurate. These tools include contextual grounding checks to evaluate AI-generated answers in real time and a Guardrails API to evaluate user prompt inputs and model responses. These tools aim to improve accuracy and filter out sensitive or harmful content. In tests, grounding checks filtered out up to 75% of hallucinations in AI model responses and blocked 85% more content when used with the Guardrails API.

Other capabilities announced at the summit include the AWS App Studio for creating AI apps from text prompts and the expansion of Amazon Q Apps for building AI apps. These updates reflect the industry trend of making generative AI tools more accessible, helpful, and accurate. For example, the AI startup Writer introduced upgrades to its platform using a graph-based approach to retrieval augmented generation (RAG) to analyze up to 10 million words when developing a chat app.

Despite the advancements in AI technology, challenges remain in adoption due to companies not knowing what they need or want to build. Providing more examples of what is possible and addressing brands’ fears of AI initiatives going wrong can help alleviate these challenges. Overall, the goal is to make generative AI applications more user-friendly, accurate, and trustworthy for a wider range of industries and applications.

Source link

Source link: https://digiday.com/media/amazon-adds-tools-for-scaling-generative-ai-applications-and-improving-accuracy-issues/

What do you think?

Leave a Reply

GIPHY App Key not set. Please check settings

Unlocking AI for Productivity: Boost Your Success

Boost productivity by unlocking AI to enhance success. #AIforProductivity

AI in 2024 Pt. 2: An Astrological Timeline of Key Developments for the Next 6 Months

Astrological timeline of AI developments in 2024 with #predictions