The bubble for artificial intelligence marketing may soon be about to burst according to a new report from research consulting firm Gartner that predicts a thirty percent reduction in AI projects by 2025.
Gartner writes in a new press release that “At least 30% of generative AI (GenAI) projects will be abandoned after proof of concept by the end of 2025, due to poor data quality, inadequate risk controls, escalating costs or unclear business value.”
Rita Sallam, the distinguished VP analyst at Gartner spoke to a group of attendees during the company’s Analytics Summit in Syndney, Australia this week and discussed the elephant in the room when it comes to AI and the material benefit to consumers that’s supposed to result in tangible profits for investors.
After last year’s hype, executives are impatient to see returns on GenAI investments, yet organizations are struggling to prove and realize value. As the scope of initiatives widen, the financial burden of developing and deploying GenAI models is increasingly felt.
Rita Sallam, Gartner VP Analysts
Sallam rings the same alarm many AI skeptics have been ringing since the marketing explosion of generative technology took hold in the public zeitgeist. When are these massive investments in large language models, commercial AI apps, coding, servers, model tuning, and personalization solutions beginning to pay off?
Perhaps, adding to the angst that some investors are starting to communicate comes as the model for GenAI remains an unpredictable technology that affords little in the way of road mapped or precedented experience for baseline profitability.
While GenAI markets a near limitless ceiling of possibilities to transform and fundamentally shift business opportunities, it also hasn’t reliably proven to be a successful model across a broad range of industry.
When it comes to dollars and cents, some companies are committing upwards of $20 million on deployment alone, not to mention initial investments in the range of billions. Simply deploying coding assistants to handle Retrieval-Augmented Generation (RAG) tools can cost up to $200,000 with a per user estimated cost of anywhere between $250 to $600 while more complex AI solutions could add anywhere between $20 million annual to a company’s bill.
With that context, it makes sense why Sallam is bullish on the notion that up to 30-percent off AI-led projects that grew root in 2021 through now, will soon be shuttering as the return-on-investment scale remains negatively unbalanced for a lot of organizations.
Sallam holds steady to the 30 percent notion even as some companies a reporting a 15 percent increase in revenue generated by GenAI projects as well as a 15 percent reduction in cost associated with GenAI project output and lastly a 22.6 percent in productivity by users whose generative AI was accompanied workload.
Ultimately, Sallam believes there will be cooling period on approved GenAI projects for some time and that organizations may come back to the table when small language models that are hyper focused on specific industries of services can show a more direct ROI, or when LLMs present a more advance use case across multiple business sectors with proven results.
