ChatGPT's most important product release to date! OpenAI "enters the B-side, fully committed to making money," directly competing with Microsoft.
After launching ChatGPT Enterprise for large-scale enterprises, OpenAI will also introduce ChatGPT Business for smaller businesses in the future. This means that OpenAI will fully expand into the enterprise user market and directly compete with its biggest funder and partner, Microsoft.
On Monday, August 28th, local time, OpenAI launched ChatGPT Enterprise, targeting large enterprises. The official claim is that this is the most powerful version of ChatGPT to date.
According to OpenAI, in addition to performing tasks such as writing emails, drafting documents, and debugging computer code like regular ChatGPT, the Enterprise version also provides enterprise-level security and privacy, as well as advanced data analysis capabilities.
Following ChatGPT Enterprise, OpenAI will also launch ChatGPT Business for smaller enterprises in the future.
This means that OpenAI will fully expand into the enterprise user market and directly compete with its biggest benefactor and partner, Microsoft. Microsoft has already provided access to ChatGPT for enterprises through its Azure OpenAI service and also offers Bing Chat Enterprise functionality to some Microsoft 365 enterprise customers.
With a large user base, OpenAI is fully focused on making money
OpenAI has already monetized ChatGPT through various means, such as offering premium subscriptions and providing API access services to enterprises.
Now, with the launch of ChatGPT Enterprise for large enterprise users and ChatGPT Business on the way for small enterprise users, this AI startup based in San Francisco is fully committed to "making money."
OpenAI's Chief Operating Officer, Brad Lightcap, refused to disclose the specific details of ChatGPT Enterprise pricing and pointed out that prices may vary depending on the needs of each enterprise. Lightcap stated that OpenAI "can work with each enterprise to find the plan that best suits them." "We have indeed attempted to build the best version of ChatGPT," Lightcap said in an interview with CNBC on Monday. "That's the team's mission: how do we build the ultimate productivity booster?"
As is well known, more powerful artificial intelligence models require more powerful computing capabilities, and large language models like GPT are no exception. Therefore, operating ChatGPT is quite costly.
According to a tweet by OpenAI CEO Sam Altman in December last year, the cost of each ChatGPT conversation is in the "single-digit cents," which suggests that providing this service to 100 million people per month could cost millions of dollars.
Lightcap stated that the development of ChatGPT Enterprise took less than a year and received assistance from over 20 companies of different sizes and industries. Its beta users include Block, Canva, and Estée Lauder.
OpenAI claims that there is significant interest from businesses in enterprise-focused ChatGPT and asserts that ChatGPT, as one of the fastest-growing consumer apps in history, has been adopted by over 80% of Fortune 500 companies.
Currently, more and more companies have found ways to use ChatGPT for programming and brainstorming. However, due to concerns about company and client data security and privacy, some companies prohibit the use of chatbots and other similar AI tools.
Lightcap told CNBC that OpenAI will first launch a version suitable for large enterprises, followed by ChatGPT Business for smaller teams, "giving us more ways to have substantive engagement with teams and fully understand deployment actions before opening up."
According to the company's blog post, OpenAI plans to "attract as many businesses as possible" in the coming weeks.
Earlier this year, Microsoft made an additional $10 billion investment in OpenAI, making it the largest AI investment of the year. In April, the startup completed a $300 million stock offering at a valuation of $27-29 billion, with participation from Sequoia Capital and Andreessen Horowitz, among others.
Just two months after its launch, ChatGPT surpassed 100 million monthly active users, breaking the record for the fastest-growing consumer application in history.
However, it is worth noting that although ChatGPT has a large user base among the general public, its position in the enterprise market is clearly lacking. Coupled with the continuous decline in global traffic for ChatGPT in recent months, expanding into the enterprise battlefield is a top priority.
How powerful is ChatGPT for large enterprises?
Media outlets have pointed out that a significant difference between ChatGPT for general consumers and ChatGPT Enterprise is that the enterprise version allows customers to input their company's data to train and customize ChatGPT for their industry and use cases. However, some features were not yet available as of Monday. ChatGPT Enterprise, powered by GPT-4, allows enterprise users to access GPT-4 with priority, without any usage limits, and at twice the speed of regular GPT-4. Moreover, the Enterprise version offers a larger input capacity, with an expanded context window of up to 32,000 tokens or approximately 25,000 words, which is four times that of the regular version.
According to an official press release from OpenAI, "We do not train our models on your business data or conversations, and our models do not learn from your specific usage." They also mentioned that customer conversation data is encrypted during transmission and while being stored. However, Lightcap stated that the company does record aggregated data on how the tool is used, including performance metadata, which is a standard practice.
ChatGPT Enterprise also provides a new management console that includes tools for managing how enterprise employees use ChatGPT, such as single sign-on, domain verification, and a dashboard with usage statistics.
Lightcap told CNBC that the biggest challenge during the development of ChatGPT Enterprise was determining the prioritization of features.
He said, "In terms of all the things that we're going to be delivering over the next few months, what I would say is that a lot of the team's focus has been on determining how to prioritize those things based on how people are using the product, what people really want, and what is empowering."
For example, the code interpreter, which was initially a feature of ChatGPT Plus, has now been renamed as "Advanced Data Analysis." Lightcap mentioned that the team questioned whether this feature belonged to the priority list for ChatGPT Enterprise, but feedback from the company led the development team to prioritize its early release.