Enterprise adoption of generative AI slows, some vendors start to help

There’s no doubt that business executives want to harness the power of generative artificial intelligence, as tools like ChatGPT continue to capture the public’s imagination. According to a study of U.S. business executives recently released by KPMG, the vast majority (60%) of respondents said that although they expect generative artificial intelligence to have a huge long-term impact, they are still far away from adopting generative artificial intelligence. AI solutions are still a year or two away.

Martin Kon, president and chief operating officer of Cohere, a Toronto-based company that provides businesses with natural language processing services powered by large-scale language models, said, “Many businesses can’t wait too long. Once they see the competition Competitors are innovating, and they are either trying to catch up or fall behind."

Not surprisingly, many service providers are scrambling to help enterprises develop and leverage generative AI capabilities.

Consulting firms are investing heavily in the field of generative artificial intelligence

For example, the world’s largest consulting firm is investing heavily in this field: Bain & Company (Bain & Company) in February this year In the middle of the year, it was the first to announce a partnership with OpenAI. Bain & Company's high-profile partnership with The Coca-Cola Company on the "Create Real Magic" campaign paid off big, as Coca-Cola highlighted in its first-quarter earnings report today.

Deloitte recently announced a new business dedicated to helping clients harness the power of generative artificial intelligence and underlying models to increase productivity and accelerate the pace of business innovation.

PricewaterhouseCoopers (PwC) also recently announced that it plans to invest US$1 billion in its U.S. operations over the next three years to adopt generative artificial intelligence technology, including partnerships with Microsoft and OpenAI Corporate cooperation. These efforts will include advising customers on how best to use generative AI while helping them build these tools.

In an interview with industry media, Mohamed Kande, vice chairman, co-leader and global consulting leader of PwC’s U.S. consulting solutions, said that many companies are happy to use generative artificial intelligence in their business , to increase productivity and other improvements, but they worry about the collateral damage that must be managed, such as data privacy risks.

That’s why a large portion of the investment will be used to build PwC’s own generative AI capabilities and expertise, he explained. "Whatever we advise customers to do when it comes to adopting and scaling technology, we do it ourselves first," he said. "Then we can say, here are all the lessons we learned from that, here's how we protect the data."

Enterprises must use generative artificial intelligence technology carefully

Rohit Gupta, founder and CEO of Auditoria, said that although large-scale language model technology has been developed for 5 to 7 years, from the perspective of enterprise deployment It seems that it is still a new technology. Auditoria is an AI-powered SaaS automation provider for corporate finance.

He explained: “Enterprises are not yet equipped with consistent methods for evaluating large language models, and the ability to quantify the ROI of such investments is still developing. Additionally, in order to harness the power of large language models, they need to make it Running on enterprise data, while many enterprises are not yet willing to open it up widely, will require additional data controls."

This means that for large enterprises, adopting generative AI is not just about consumption. Log in to the Internet and use ChatGPT as usual.

Kande said companies must not only understand how to manage the data they want to use, but also understand the risks within the enterprise. "We tell them, it's not going to happen in one day," he said. On the other hand, he noted, not all use cases are risky. “Some of it actually helps increase productivity without causing any collateral damage,” he said. Drayton Wade, director of product operations and strategy at Kognitos, an AI automation platform, said that it is generative AI that The novelty and dispersion of it have stopped many people from moving forward. However, it is and can be used safely in today's enterprises, especially when it comes to automation.

"When combined with a deterministic logic system, it can be used immediately for safety and productivity," he said. He added that business executives should look for AI-based generative platforms that The platform features human review steps, full auditability (in natural language) and a privacy system.

As generative AI providers look to capitalize on the enterprise market, even ChatGPT looks set to get in on the action.

OpenAI company said in a blog post recently, "We are developing a new ChatGPT Business subscription for professionals who need more control over their data, as well as enterprises looking to manage their end users. ChatGPT Business will follow Our API’s data usage policy means that by default, end-user data will not be used to train our models. We plan to launch ChatGPT Business in the coming months."

But OpenAI’s competitor Cohere, which specializes in customizing large language models, believes that OpenAI’s products cannot meet the needs of enterprises.

I'm sure it's going to be a great product, said Kon, COO of Cohen. But he cautioned that for mission-critical enterprise use cases, businesses won't want to use what everyone is using. common standard tools because they need to have a competitive advantage. Therefore, by definition, enterprises need to develop these products based on their own large-scale language models, operate in their own data environment, and use their own proprietary data. ”

Overcoming Fear and Moving Toward an Artificial Intelligence Future

While many top enterprise companies have already boarded the generative AI train, such as Walmart, which recently said it is working on OpenAI’s GPT- 4, but it must overcome the fear that comes with the excitement.

PwC’s Kande said, “Italy’s ban on ChatGPT was motivated by concerns about how to protect data. We personally think this technology is available - don't be afraid, but manage the risks. He added that PwC needs to develop its own generative AI systems to teach clients lessons about delivering results. "This changes the nature of our discussions with clients because This is not just a discussion about artificial intelligence. ”

The above is the detailed content of Enterprise adoption of generative AI slows, some vendors start to help. For more information, please follow other related articles on the PHP Chinese website!