OpenAI has introduced ChatGPT Enterprise, a business-centric variant of ChatGPT, incorporating enhanced features and privacy measures.
Similar to its consumer versions, the business-oriented ChatGPT Enterprise allows users to input a text prompt and get a response generated by OpenAI’s most advanced generative AI model, GPT-4.
The enterprise version also includes features like data encryption and the assurance that OpenAI will not use client data to improve its technology.
Additionally, the enterprise variant handles a longer context window, enabling users to input longer prompts – particularly useful for data and document analysis.
OpenAI’s ChatGPT Enterprise signals a continuation of the company’s efforts to monetize its models, following on from ChatGPT Plus and API offerings.
Running models like ChatGPT demands vast computing resources, making it a costly endeavor for AI companies.
OpenAI’s COO, Brad Lightcap, refrained from specifying the pricing details for ChatGPT Enterprise, stating that costs could be tailored to meet the individual needs of each business.
“We’ve really tried to build the best version of ChatGPT,” said Lightcap. “That was the mandate for the team: How do we build something that’s the ultimate productivity enhancer?”
Brad Lightcap also revealed that OpenAI collaborated with over 20 companies of various sizes and with varying degrees of AI experience to test ChatGPT Enterprise. Companies such as Estée Lauder, Canva Inc., and Zapier Inc. are among its early adopters, using the tool for tasks ranging from coding to creative work and business decision-making.
According to OpenAI, ChatGPT has been adopted by some 80% of Fortune 500 companies since its initial rollout in November.
Despite growing usage, some organizations have hesitated to adopt ChatGPT and other AI tools due to data security and privacy concerns. Many businesses have turned to open-source AI models that can be trained and optimized with their own data without extra costs or risks.
Addressing these issues, OpenAI clarified that it will not use data or prompts from companies utilizing ChatGPT Enterprise to train its AI models.
OpenAI currently trains its models on data submitted via the ChatGPT online interface or mobile apps, although users can opt-out, and it does not use data transmitted through its API for training purposes.
Companies have also expressed concerns about large language models (LLMs) like ChatGPT generating inaccurate or fabricated information. There’s not much businesses can do about that, and they’ll have to exercise due diligence when utilizing the AI’s outputs.
Here is the complete feature list of ChatGPT Enterprise:
- Unrestricted use of GPT-4 (no limitations on usage)
- Enhanced GPT-4 speed performance (twice as fast)
- Unlimited utilization of advanced data analysis (previously named Code Interpreter)
- Extended 32k token context windows for quadrupled input length, files, or sequential interactions
- Collaborative chat templates for streamlined company workflows
- Complimentary API credits for custom OpenAI solutions