OpenAI Unveils GPT-4 Turbo and Copyright Shield for Enterprise Users
OpenAI launches a faster GPT-4 Turbo model and introduces a Copyright Shield to protect enterprise and API customers from infringement claims. The announcements come alongside new customizable bots and a planned app store during the company's first developer conference.
OpenAI hosts its first developer conference in San Francisco and unveils GPT-4 Turbo, a faster foundational model that accepts prompts up to 128,000 tokens. The company also introduces customizable, low-code bots called GPTs and plans to launch an app store with revenue sharing for developers. Microsoft CEO Satya Nadella makes a guest appearance to praise the ongoing partnership between the two tech giants.
To address growing legal concerns in the AI industry, OpenAI announces a Copyright Shield for its API and ChatGPT Enterprise customers. This indemnification plan means OpenAI steps in to defend users and pay any costs if they face copyright infringement claims over AI-generated output. CEO Sam Altman emphasizes that the company never trains its models on data submitted through the API or ChatGPT Enterprise.
The updated model and legal protections arrive as OpenAI's API user base grows to over two million developers integrating models like GPT-4, DALL·E, and Whisper into their applications. These moves aim to make AI tools more accessible and safer for businesses to adopt at scale. The new offerings position OpenAI to compete more aggressively in the rapidly evolving enterprise AI market.