OpenAI<\/a>, Anthropic and Google are all doubling down on courting business customers in the next year. During a recent lunch with media editors in New York, OpenAI CEO Sam Altman said developing AI systems for companies could be a $100 billion market.<\/p>\nAll this is happening against the backdrop of unprecedented tech investment in everything from chips, to data centers, to energy sources.<\/p>\n
Whether these investments can be justified will be determined by companies\u2019 ability to figure out how to use AI to boost revenue, fatten margins or speed innovation. Failing that, the infrastructure build-out could trigger the kind of crash reminiscent of the dot-com bust in the early 2000s, some experts say.<\/p>\n
THE ‘EASY’ BUTTON<\/p>\n
Soon after ChatGPT\u2019s launch, companies worldwide created task forces dedicated to finding ways to embrace generative AI, a type of AI that can create original content like essays, software code and images through text prompts.<\/p>\n
One well-known issue with AI models is their tendency to please the user. This bias \u2013 what\u2019s called \u201csycophancy\u201d \u2013 encourages users to chat more, but can impair the model\u2019s ability to give better advice.<\/p>\n
CellarTracker ran into this problem with its wine-recommendation feature, \u2060built on top of OpenAI\u2019s technology, CEO LeVine said. The chatbot performed well enough when asked for general recommendations. But when asked about \u200cspecific vintages, the chatbot remained positive \u2013 even if all signals showed a person was highly unlikely to enjoy them.<\/p>\n
\u201cWe had to bend over backwards to get the models (any model) to be critical and suggest there are wines I might not like,\u201d LeVine said.<\/p>\n
Part of the solution was designing prompts that gave the model permission to say no.<\/p>\n
Companies have also struggled with AI\u2019s lack of consistency.<\/p>\n
Jeremy Nielsen, general manager at North American railroad service provider Cando Rail and Terminals, said the company recently tested an AI chatbot for employees to study internal safety reports and training materials.<\/p>\n
But Cando ran into a surprising \u200dstumbling block: the models couldn\u2019t consistently and correctly summarize the Canadian Rail Operating Rules, a roughly 100-page document that lays out the safety standards for the industry.<\/p>\n
Sometimes the models forgot or misinterpreted the rules; other times they invented them from whole cloth. AI researchers say models often struggle to recall what appears in the middle of a long document.<\/p>\n
Cando has dropped the project for now, but is testing other ideas. So far the company has spent $300,000 on developing AI products.<\/p>\n
\u201cWe all thought it\u2019d be the easy button,\u201d Nielsen said. \u201cAnd that\u2019s just not what happened.\u201d<\/p>\n
HUMANS MAKE A COMEBACK<\/p>\n
Human-staffed call centers and customer service were supposed to be heavily disrupted by AI, but companies quickly learned there are limits to the amount of human interaction that can be delegated to chatbots.<\/p>\n
In early 2024, Swedish payments company Klarna rolled out an OpenAI-powered customer service agent that it \u200bsaid could do the work of 700 full-time customer service agents.<\/p>\n
In 2025, however, CEO Sebastian Siemiathowski was forced to dial that back and acknowledge that some customers preferred to talk with humans.<\/p>\n
Siemiathowski said AI is reliable on simple tasks and can now do the work of about 850 agents, but more complex issues quickly get referred to human agents.<\/p>\n
For 2026, Klarna is \u200cfocused on building its second-generation AI chatbot, which it hopes to ship soon, but human beings will remain a big part of the mix.<\/p>\n
\u201cIf you want to stay customer-obsessed, you can’t rely [entirely] on AI,\u201d he said.<\/p>\n
Similarly, U.S. telecommunications giant Verizon is leaning back into human customer service agents in 2026 after attempts to delegate calls to AI.<\/p>\n
\u201cI think 40% of consumers like the idea of still talking to a human, and they’re frustrated that they can’t get to a human agent,\u201d said Ivan Berg, who leads Verizon\u2019s AI-driven efforts to enhance service operations for business customers, in a Reuters interview this fall.<\/p>\n
The company, which has about 2,000 frontline customer service agents, still uses AI to screen calls, get information on customers, and direct them to either self-service systems or to human agents.<\/p>\n
Using AI to handle routine questions frees up agents to handle complex issues and try new things, such as making outbound calls and doing sales.<\/p>\n
\u201cEmpathy is probably the key thing that’s holding us from having AI agents talk to customers holistically right now,\u201d Berg said.<\/p>\n
Shashi Upadhyay, president of product, engineering and AI at customer-service platform Zendesk, says AI excels in three areas: writing, coding and chatting. Zendesk\u2019s clients rely on generative AI to handle between 50% and 80% of their customer-support requests. But, he said, the idea that generative AI can do everything is \u201coversold.\u201d<\/p>\n
THE \u2018JAGGED FRONTIER\u2019<\/p>\n
Large language models are rapidly conquering complex tasks in math and coding, but can still fail at comparatively trivial tasks. Researchers call this contradiction in capabilities the \u201cjagged frontier\u201d of AI.<\/p>\n
\u201cIt might be a Ferrari in math but \u2060a donkey at putting things in your calendar,\u201d said Anastasios Angelopoulos, the CEO and cofounder of LMArena, a popular benchmarking tool.<\/p>\n
Seemingly small issues can unexpectedly trip up AI systems.<\/p>\n
Many financial firms rely on data compiled from \u200ba broad range of sources, all of which can be formatted very differently. These differences might prompt an AI tool to \u201cread patterns that don\u2019t exist,\u201d said Clark Shafer, director at advisory firm Alpha Financial Markets Consulting.<\/p>\n
Many companies are now \u200blooking into the potentially expensive, lengthy and complex process of reformatting their data to take advantage of AI, Shafer said.<\/p>\n
Dutch technology investment group Prosus says one of its in-house AI agents is meant to answer questions about its portfolio, similar to what the group\u2019s data analysts on staff already do.<\/p>\n
Theoretically, an employee could ask how often a Prosus-backed food-delivery firm was late to deliver sushi orders in Berlin last week.<\/p>\n
But for now, the tool doesn\u2019t always understand what neighborhoods are part of Berlin or what \u201clast week\u201d means, said Euro Beinat, head of AI for Prosus.<\/p>\n
\u201cPeople thought AI was magic. It’s not magic,\u201d Beinat \u200dsaid. \u201cThere’s a lot of knowledge that needs to be encoded in these tools to work well.\u201d<\/p>\n
MORE HANDHOLDING<\/p>\n
OpenAI \u2060is working on a new product for businesses and recently created internal teams, such as the Forward Deployed Engineering team, to work directly with clients to help them use OpenAI\u2019s technology to tackle specific problems, a spokesperson said.<\/p>\n
\u201cWhere we do see failure is people that jump in too big, they find that billion-dollar problem\u2014that’s going to take a few years,\u201d said Ashley Kramer, OpenAI\u2019s head of revenue, during an onstage interview at Reuters Momentum AI conference in November.<\/p>\n
Specifically, OpenAI is working with companies to find areas where AI can have a \u201chigh impact but maybe low lift at first,\u201d said Kramer.<\/p>\n
Rival AI lab Anthropic, which draws 80% of its revenue from business customers, is hiring \u201capplied AI\u201d experts who \u2060will embed with companies.<\/p>\n
For AI companies to succeed, they will have to view themselves as \u201cpartners and educators, rather than just deployers of technology,\u201d said Mike Krieger, Anthropic\u2019s head of product, in an interview earlier this year.<\/p>\n
An increasing number of startups, many founded by former OpenAI employees, are developing AI tools for specific sectors such as financial services or legal. These founders say companies will benefit from specialized models more than general-purpose or consumer tools like ChatGPT.<\/p>\n
It\u2019s a playbook \u200cthat Writer, a San Francisco\u2013based AI application startup, has been adopting. The company, which is now building AI agents for finance and marketing teams at large firms such as Vanguard and Prudential, puts its engineers on calls directly with clients to understand their workflows and co-build the agents.<\/p>\n
\u201cCompanies need more handholding in actually making \u200cAI tools useful for them,\u201d said May Habib, CEO of Writer.<\/p>\n
(Reporting by Deepa Seetharaman and Krystal Hu in San Francisco and Supantha Mukherjee in Stockholm. Editing by Kenneth Li and Michael Learmonth.)<\/p>\n","protected":false},"excerpt":{"rendered":"By Deepa Seetharaman, Supantha Mukherjee and Krystal Hu SAN FRANCISCO\/STOCKHOLM, Dec 16 – Last spring, CellarTracker, a wine-collection…\n","protected":false},"author":2,"featured_media":194807,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[220,218,219,104046,104045,4810,6113,104047,104044,1223,61,60,7601,80],"class_list":{"0":"post-194806","1":"post","2":"type-post","3":"status-publish","4":"format-standard","5":"has-post-thumbnail","7":"category-artificial-intelligence","8":"tag-ai","9":"tag-artificial-intelligence","10":"tag-artificialintelligence","11":"tag-business-customers","12":"tag-cellartracker","13":"tag-chatbot","14":"tag-customer-service","15":"tag-eric-levine","16":"tag-forrester-research","17":"tag-generative-ai","18":"tag-ie","19":"tag-ireland","20":"tag-tech-companies","21":"tag-technology"},"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/194806","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/comments?post=194806"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/posts\/194806\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media\/194807"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/media?parent=194806"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/categories?post=194806"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/ie\/wp-json\/wp\/v2\/tags?post=194806"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}