Microsoft Puts AI Chat To Work With Bing Chat Enterprise

By Melody Brue, Patrick Moorhead - August 14, 2023

Microsoft CEO Satya Nadella announced in February, “The [AI] race starts today.” That proclamation accompanied Microsoft’s announcement of AI-powered Bing Search, which came on the heels of the company’s $10 billion investment in OpenAI, an add-on to multiple rounds of funding that began with an initial $1 billion in 2019. Since then, every tech company has incorporated some form of AI into its products—or at least its product roadmap—either to seize a real opportunity or simply out of fear of becoming obsolete. This frothiness reminds me of prior inflections where investors put pressure on the board, which applied pressure to the C-Suite, which applied pressure to the entire organization. It has been like this with the introduction of the Web, e-commerce, mobile, social, AI and now generative AI (GAI) based on foundation models.

Among the tech giants, Microsoft has set a remarkable pace of AI innovation this year and has emphasized the transformative potential of AI in reshaping every aspect of business. Yet, since the introduction of ChatGPT and other AI-powered chatbots, many companies, including Samsung, JPMorgan Chase and Verizon, have banned employees from using them due to the potential for sensitive data leaks. Even Google, which has implemented Anthropic’s “Claude” AI assistant in some of its products, has discouraged employees from using the chatbot with company information.

At its annual Inspire conference this week, Microsoft announced Bing Chat Enterprise, an AI-powered chat platform for business that addresses a gap in the market for GAI with commercial data protection.

The AI race has become more than a contest of speed. It signifies what could be the most transformational technological shift in history and is poised to change how work is done. In this analysis, I’ll look at Microsoft’s generative AI advancements and their potential to drastically impact workers and productivity.

Microsoft Chairman and CEO Satya Nadella at Microsoft headquarters in Redmond, Washington
Dan Delong for Microsoft

The consumer phenomenon of ChatGPT

ChatGPT hit the market in November 2022 and became the most quickly adopted consumer app in history. However, the large language model (LLM) underlying ChatGPT was current only through 2021, so getting more up-to-date answers from the chatbot was a dead end. Microsoft offered a solution for this by harnessing ChatGPT, combining it with current information from the Web and calling it Bing Chat. At that time, the solution was focused on consumers, not enterprises.

Forbes Daily: Get our best stories, exclusive reporting and essential analysis of the day’s news in your inbox every weekday.

When Microsoft first announced Bing Chat powered by ChatGPT in February, it met with tremendous excitement from the market. Microsoft says the feedback from many IT pros, security decision makers and others was that they were excited about what generative AI could bring to their organizations. However, they were also concerned about how GAI solutions popular with consumers would use sensitive data in enterprise settings. While task automation and the corresponding boost in productivity are promising for organizations, the risk of disclosing private data and the potential for generating malicious code and false or harmful content cannot be ignored.

Bing Chat, but make it for work

Companies want to leverage their enterprise data using AI. The challenge is to bring that data into a good AI model without inadvertently putting the business at risk. As with the consumer version of Bing Chat, the new Enterprise version combines GAI chatbot functionality with web data from Bing search. The big differences are commercial-grade data protection and the injection of proprietary enterprise data from the organization using the product.

Putting all of this together enables Bing Chat Enterprise to provide detailed and relevant responses based on current information. The commercial data protection ensures that sensitive business data remains secure within the organization. Even better, each answer comes with a citation so users can validate internal and external sources. This helps validate the responses and develop trust. That said, enterprise users need to be aware that, although their company data is secure, chatbot responses may still include factual errors. To minimize the likelihood of inaccuracies or “hallucinations,” users should compose precise and well-structured prompts and verify citations. The quality of what is fed into the chatbot—the prompt—will often influence the quality of the response. I’d like to see Microsoft provide “prompt engineering” tips, similar to how Adobe has done with Adobe Firefly.

With commercial-grade security, users should be able to confidently inquire about internal documents, get assistance writing code, create product sheets or even ask for a SWOT analysis. These are some examples, and Microsoft demoed a few of them, but we have not yet scratched the surface of the productivity gains that enterprise AI will be able to produce. As models are more finely tuned and workers become more comfortable using prompts to produce needed outcomes, tools like Bing Chat Enterprise will significantly impact working styles and productivity.

Bing Chat Enterprise is currently in preview and will be accessible on all Bing Chat–supported platforms, including Chat and the Microsoft Edge sidebar. It will eventually be available in Windows Copilot, too. It is provided at no extra cost to customers who have subscribed to Microsoft 365 E3, E5, Business Standard and Business Premium. It will eventually be offered as a standalone service for $5 per monthly user.

Microsoft also announced Visual Search in Bing Chat, allowing users to search with images. Using OpenAI’s GPT-4 model, Visual Search in Bing Chat enables users to upload images and conduct web searches for relevant content. Bing can comprehend the context of an image, interpret its content and respond to related queries.

Microsoft Copilot

Microsoft 365 Copilot gets a price tag

Microsoft’s 365 Copilot subscription service adds AI to the company’s popular products such as Word, Excel, PowerPoint and Teams. The company announced this week that Microsoft will price 365 Copilot at $30 per user per month when it becomes generally available. Microsoft 365 Copilot was introduced earlier this year, intended as the ultimate copilot for work. It harnesses the power of LLMs, along with data from the Microsoft Graph and the Microsoft 365 suite of apps. Copilot offers personalized assistance to help users generate insights and complete tasks more efficiently.

I am certainly not the only one champing at the bit for Microsoft 365 Copilot to be generally available. I can’t say I’m patiently waiting, but I appreciate that Microsoft has shared pricing with customers to help them plan for when it’s time to implement. Some of the use cases that I know will be popular among Moor Insights & Strategy analysts will be the ability to create PowerPoints out of Word documents, generate documents out of presentations and get better cross-company access to intelligence.

I believe Microsoft has an opportunity to better highlight the incremental value beyond $30—by showcasing the benefits of Microsoft 365 Copilot beyond what is offered in Bing Chat Enterprise. The value of Microsoft 365 Copilot is that users can pull in all of their enterprise data and web search data like they can with Bing Chat Enterprise, but the functionality is also integrated directly into the Microsoft 365 apps. This means there is no toggling to access tools, and that this powerful new GAI is embedded in contexts that are already familiar to users. Want a strong ROI that just about every corporate professional will understand? How about this: “No learning curve to create a working version of a PowerPoint deck extracted directly from company data and an email from your boss—in three minutes.” Microsoft needs to be specific about how these solutions build on each other and add incremental value for users.

Wall Street bells were Binging with the news

Additional payments through recurring subscriptions of $30 per month could increase monthly prices for enterprise customers by as much as 83%, and Microsoft’s investors reacted positively to this news. On Tuesday, Microsoft’s shares reached an all-time high following the company’s pricing announcement. The stock experienced a significant 4% increase, ending the day above $359. Microsoft’s stock has seen a 50% rise since the beginning of the year.

Executives at Microsoft said that pricing for the new offering was determined by speaking with customers and establishing the value customers were getting from it. These models are costly to run, and customers recognize the business value created by GAI tools, particularly those equipped with enterprise security. Also, the additional $30-per-seat license doesn’t have to cover every seat of an enterprise subscription. This is a productivity tool that companies can assign to organizational positions, which should make the price tag more palatable for organizations that need it only in certain departments or want to test it out with a smaller group first.

Wrapping up

Bing Chat Enterprise should come as a surprise to nobody. Microsoft laid the groundwork for enterprise GAI at Microsoft Build in June with updated features for copilots catering to a wide range of business users. These include Dynamics 365 Copilot, Microsoft 365 Copilot and Copilot for Power Platform.

I believe Microsoft’s dedication to enterprise-grade solutions has remained unwavering during the ongoing AI revolution. Organizations’ potential to reshape how people work using AI is incredibly promising. Once Microsoft 365 Copilot is widely available, I look forward to seeing how real-world applications change the future of work.

AI will revolutionize our work, freeing time by handling routine tasks and allowing employees to focus on important strategic and creative projects. With the increased use of AI-driven automation, there is a widespread fear that specific job roles may become obsolete, leading to potential job displacement for some workers. This may be true and remains to be seen. However, AI will likely create new job opportunities and demand specialized skills to develop, maintain and oversee AI systems. Additionally, as AI assistants and chatbots reshape the workforce, adapting to new technologies will require upskilling and reskilling. There is a real opportunity to build training into AI products so employees can extract the most value.

As I said, the AI race has become not just a contest of speed. However, the rate at which AI will change how we work will likely earn a place in the history books. Future generations can ask Bing Chat about it—and hopefully get accurate answers grounded in detailed context.

Note: This analysis contains content from Moor Insights & Strategy CEO and Chief Analyst Patrick Moorhead.

Melody Brue
+ posts

Mel Brue is vice president and principal analyst covering modern work and financial services. Mel has more than 25 years of real tech industry experience in marketing, business development, and communications across various disciplines, both in-house and at agencies, with companies ranging from start-ups to global brands. She has built a unique specialty working in technology and highly regulated spaces, such as mobile payments and finance, gaming, automotive, wine and spirits, and mobile content, ensuring initiatives address the needs of customers, employees, lobbyists and legislators, as well as shareholders. 

Patrick Moorhead
+ posts

Patrick founded the firm based on his real-world world technology experiences with the understanding of what he wasn’t getting from analysts and consultants. Ten years later, Patrick is ranked #1 among technology industry analysts in terms of “power” (ARInsights)  in “press citations” (Apollo Research). Moorhead is a contributor at Forbes and frequently appears on CNBC. He is a broad-based analyst covering a wide variety of topics including the cloud, enterprise SaaS, collaboration, client computing, and semiconductors. He has 30 years of experience including 15 years of executive experience at high tech companies (NCR, AT&T, Compaq, now HP, and AMD) leading strategy, product management, product marketing, and corporate marketing, including three industry board appointments.