Executives say they don’t know how to make AI profitable – 01/18/2024 – Tech

Bright banners herald the promise of artificial intelligence along Davos’ main promenade, but executives gathered at the World Economic Forum say they are struggling to figure out how to make a return on their investment in the technology.

Several company CEOs gathered at this year’s forum told Reuters that current generative artificial intelligence technology still has a lot to prove.

The president of cloud computing systems security company Cloudflare, Matthew Prince, told Reuters that the next few months could even feel like an “AI letdown.”

“Everyone thinks, yes, I can create these cool demos, but where’s the real value?” he said, echoing a theme among business leaders at the Davos event.

ChatGPT’s rapid growth is, in some ways, an exception. In the first two months since its launch in November 2022, the chatbot reached around 100 million users, making it one of the fastest growing apps in history.

The chatbot brought so-called generative AI to consumers’ fingertips, allowing people to write what they want in a search field and get a poem, a school essay or collect information on various topics.

The product has also proven to be a good contributor to developing ideas in “low-risk, non-business-critical use cases,” said Victor Riparbelli, chief executive of AI video generation startup Synthesia.

But “enterprise applications are definitely not really ready” for this chat-based AI, the executive said.

One of the problems cited by Riparbelli is that there is no clear path to ending so-called “hallucinations”, or content with misinformation generated by AI.

While computer scientists have developed methods to narrow down where chatbots can extract answers, business leaders may not want to take that risk.

Other concerns, said Ana Paula Assis, president of IBM for Europe, the Middle East and Africa, are preventing the chatbot’s AI from reproducing human biases.

“Customers are still very concerned about how to bring these solutions within the confines of regulations and compliance,” she said.

Chinese Premier Li Qiang said in Davos that AI needs to serve the common good but must be properly governed as it “poses risks to security and our ethics” and Chinese President Xi Jinping wants United Nations play a central role in discussions on AI, said UN Secretary-General António Guterres on Wednesday (17).

Meanwhile, about 90% of a sample of 1,400 C-suite executives said they expect generative AI to take a step beyond its recent popularity or are doing only limited experiments and pilots, according to survey results published by consultancy BCG.

But big tech companies, including Microsoft, Alphabet and Amazon.com, have forged ahead, courting thousands of companies to try newer AI systems.

Microsoft Chief Executive Satya Nadella said at a company event in Davos on Wednesday that AI is poised to boost productivity and potentially accelerate science itself.

However, companies’ revenue and profits from recent efforts are still unclear.

“BE REALISTIC ABOUT AI”

While a sign at Davos invited people to “get real about AI,” efforts to find a market for it have developers considering a variety of places.

Cohere, a top-tier enterprise-focused AI startup, sees helping sellers as a path to revenue.

“The idea is that it’s on the sales side and that it makes sales teams more productive,” Cohere chief executive Aidan Gomez told Reuters. The hope would be to “help them make more contacts, more follow-ups and automate a lot of that process.”

On the other hand, medicine is more complicated. While speeding up note-taking for doctors is a task worthy of AI, automating the medical profession is not, as doing so could risk lives, Gomez said.

“We should focus on helping humans, not replacing doctors and having a chatbot doctor,” said Gomez.

Novartis Chief Executive Vasant Narasimhan said the drugmaker is working with Microsoft to more widely deploy AI to provide samples to staff that submit 20,000 to 30,000 responses to regulator queries a year. The “next opportunity,” he said at the Microsoft event, would be AI for drug design.

Tejpreet Chopra, chief executive of BLP Group, a large wind and solar energy operator in India, told Reuters that the company is ready to incorporate AI chat technology, “but only for internal use, to write a good English, not for content”.

ELECTIONS

Elections are a high-risk area for AI companies as voters around the world head to the polls in 2024.

Regarding the use of AI in disinformation campaigns, Gomez said Cohere’s policies prohibit impersonation, while Riparbelli said Synthesia does not allow customers to create political content through its AI video platform.

OpenAI, which also prohibits abusive impersonation through its technology, said Monday that it is working with the U.S. National Association of Secretaries of State and will begin directing users to CanIVote.org for election-related questions.

Understanding how content is created is a key concern among companies and policymakers, said Arati Prabhakar, director of the White House Office of Science and Technology Policy.

“If (people) see a video or an image, they should be able to tell whether it was generated by AI or by humans,” Prabhakar told Reuters.

For Srini Pallia, an executive at technology consulting and services firm Wipro, the buzz about AI at Davos is loud and clear, filling the void left by cryptocurrencies.

“It’s AI, AI and more AI,” Pallia said.