Posted on 08/19/2025 4:42:33 PM PDT by libh8er
A new report from MIT’s NADA initiative reveals a sobering reality for corporate generative AI adoption: 95% of pilot programs are failing to deliver meaningful financial impact. Titled The GenAI Divide: State of AI in Business 2025, the study is based on 150 interviews with business leaders, a survey of 350 employees, and an analysis of 300 public AI deployments. It highlights a stark divide between successful AI integrations and stalled initiatives that offer little to no return on investment [1].
According to the research, only 5% of AI pilots result in rapid revenue acceleration. The majority of projects remain stuck at the initial phase, unable to scale beyond experimental stages. Aditya Challapally, lead author of the report and head of the Connected AI group at MIT Media Lab, attributes this failure not to the quality of AI models themselves, but to a "learning gap" between the tools and the organizations implementing them. Unlike consumer tools like ChatGPT, which offer flexibility for individual users, enterprise AI tools struggle to adapt to internal workflows unless specifically tailored [1].
The report further finds that more than half of generative AI budgets are allocated to sales and marketing, despite the most significant returns being observed in back-office automation. These include reductions in business process outsourcing, lower external agency costs, and improved operational efficiency. This misalignment suggests a lack of strategic clarity in how companies are investing in AI [1].
(Excerpt) Read more at ainvest.com ...
That is to be expected.
My mind goes back to the .com bubble.
This AI stuff is going to be one big flop. it was have its uses of course but the boomers are going to look like fools.
What they found was that the use of AI actually decreased programmer productivity by 18%, but that the coders believed their productivity had increased by 20%.
When you call Verizon for support, you get an annoying robot voice that tells you they now use AI for advanced customer support. It is entirely and wholly useless. All it does is throw another roadblock between you and somebody who can really help you solve a problem.
As my wife found out this evening.
(Lost 1:15, only useful result was to “get an appointment scheduled tomorrow “.
At the local store.
With a person.
Am I being too cynical when I imagine that companies intend for their phone assistant automation to be so useless and frustrating that customers will just give up and live with their problems?
At the very least, I suspect that the frustration is intended to weed out callers with minor problems so that the expensive human help is minimized.
AI is only as good at coding as the programmer prompting it. If you have crappy programmers, you’ll have crappy code. The only difference is they do it faster.
No, not too jaded at all. It’s infuriating. The new Verizon IVR greeting is the worst — they tell you all about your superior customer experience with AI. I give the a solid zero out of ten.
“According to the research, only 5% of AI pilots result in rapid revenue acceleration.”
and 100% of the successful 5% made their money by convincing the other 95% that they could make money by buying AI from the successful 5% ...
The title of this article is misleading.
AI itself isn’t the issue—the problem is that non-AI companies integrating AI systems misallocate budgets to sales/marketing and rely on poor adoption strategies.
The report shows that good financial returns come from using external, proven AI systems and back-office efficiency, not proprietary lab projects or hype-driven marketing.
This article is about non-AI companies trying to implement AI into their operations. And they are doing it poorly, by trying to set up their own AI labs to develop proprietary AI systems.
Real AI companies like Nvidia, AMD, Microsoft, Amazon, Alphabet, and others are among the most profitable companies in history. The dot-com bubble was built on IPOs for companies with no profits and often no real business—just ideas and a website.
I can submit the 10-Qs of, say, 5 competing companies to AI and have it compare each and every financial line item and then summarize the results, give me the reasons why they differ, and match the final conclusions against stock market perceptions.
I have to check the details carefully, of course, but performing this level of analysis would typically take a rather large team of extremely competent (and expensive) analysts. AI does it in about 15 seconds and the cost is $20 to ChatGPT and ~$8 to Grok, monthly.
I am considering buying my own AI-capable computer to develop and train my own AI system unconnected from the internet. My own AI system would run locally for privacy and control, but it would rest on a more powerful foundation like ChatGPT. Think of it like having a home office computer that does most tasks, but when I need deep research or advanced processing, it connects to ChatGPT’s larger engine. That way I get both independence and access to enterprise-level intelligence.
In both cases there is a core of enterprises that will make serious use of the technology, and bring about iinfluential and lasting changes, surrounded by a sea of projects that make ineffectual use of the technology with a great show of modernity, making for a rush to investment in companies of limited potential. I’d say they have a great deal in common.
I've talked to professionals who are already realizing big advantages from AI, however that does not mean that the advantages realized will even remotely justify the vast sums that are being poured into the technology.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.