Agentic AI faces a reckoning as enterprises demand real business value
Read the Portuguese translation →
Author: Dewain Robinson - The Agent Dude · Watch on YouTube
Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.
Original language: English
Original video: Agentic AI faces a reckoning as enterprises demand real business value · Channel: Dewain Robinson - The Agent Dude · 03:10 · Watch on YouTube
Topics: AI agents, Enterprise AI applications
Summary
Section titled “Summary”The video discusses Gartner’s prediction that 40% of agentic AI projects will be abandoned by 2027, arguing that this reflects weak business value and ROI rather than a decline in AI overall. Organizations are getting better at distinguishing useful AI projects from pilots and products that adopt the technology mainly to follow a trend. The host warns about “agent washing” and expects fewer companies to remain in the enterprise AI space as buyers demand solutions that address real problems and deliver measurable value.
Key points
Section titled “Key points”- Gartner predicts that 40% of agentic AI projects will die by 2027.
- The expected failures are linked to projects lacking clear business value or return on investment.
- Some AI pilots have repeatedly failed because technology was implemented for its own sake.
- Organizations are improving at evaluating the value of AI projects.
- The host describes “agent washing” as adding an agent without ensuring it provides value.
- Useful AI should solve new problems rather than merely add a fancy interface to an existing API.
- The video anticipates a reduction in the number of companies claiming to be AI companies.
- Buyers are advised to assess whether an AI product provides genuine value and whether its provider is likely to last.
Chapters
Section titled “Chapters”| Start | Chapter | Summary |
|---|---|---|
| 00:01 | Gartner’s forecast for agentic AI | The host introduces Gartner’s prediction that 40% of agentic AI projects will die by 2027 and clarifies that this does not mean AI itself is declining. |
| 00:33 | Business value and project evaluation | The discussion links project failures to weak business value and explains that organizations are getting better at judging AI initiatives. |
| 01:33 | Agent washing and real problem-solving | The host distinguishes trend-driven agent adoption from AI that solves new problems, and cites an expected reduction in companies in the AI space. |
| 02:37 | What buyers should look for | The video urges buyers to verify that AI offerings provide genuine value and come from organizations likely to remain in the market. |
Quotes
Section titled “Quotes”Gartner’s saying that 40% of agentic AI projects die by 2027. — 00:01
there are a lot of projects that are going on within organizations that don’t actually have good business value or good return on investment — 00:33
what we refer to as agent washing — 01:33
actually solve new problems, not just put a fancy UI on top of an existing API. — 02:04
ask yourself, is this organization truly an AI organization — 02:37
Concepts and entities
Section titled “Concepts and entities”Gartner, agentic AI, agent washing, return on investment
Full transcript (original language)
Section titled “Full transcript (original language)”00:01 Welcome to the Cloud Wars AI Minute. I’m your host Dwayne Robinson, and we’re going to go over the things that your boss expects you to know about what’s happening in the world of AI. And the one I want to talk about today is that we’re not we’re starting to see some interesting analysis coming out of organizations like Gartner. So, Gartner’s saying that 40% of agentic AI projects die by 2027. Now, while that sounds pretty daunting, it’s not that they’re actually saying that that we’re going to see a reduction in AI, it’s
00:33 just that there are a lot of projects that are going on within organizations that don’t actually have good business value or good return on investment, and we’re starting to see organizations be able to determine what are good value AI projects versus poor uh projects. And we already started to see a lot of pilots that were just failing over and over and over because people were trying to implement technology just for technology’s sake. So, I think we’re
01:03 going to definitely see an improvement across the organization around the world and just being able to get better at determining whether or not there is a value return on these different AI projects that people are doing across the board. And what you’re also seeing is that right now we’ve got about 130-ish or so uh as Gartner counts as legitimate real uh organizations doing AI things.
01:33 And so, what what we want to see is that a separation between organizations that are just basically doing what we refer to as agent washing, which is just taking and putting in an agent, whether that agent is actually doing anything of value or not within their organization or as a product, they’re just following that trend. Where what we really want to see is agents or AI technology that’s
02:04 coming in to actually solve new problems, not just put a fancy UI on top of an existing API. So, those type of things we should continue to see organizations getting better at this, but it will mean we will see a reduction in the number of players in this space over time. Matter of fact, what we’re expecting is that if we really look at this and we follow what Gartner is saying, they believe that we’re going to see about a 33% reduction
02:37 in the number of companies that are actually playing in the AI space or enterprise software companies that are claiming to be AI type of companies. And a lot of it has to do with just not being able to justify the value behind what they’re doing and doing this agent washing type of capabilities. So, make sure you’re paying attention to anything that you’re looking to purchase as you go into this whole AI space and ask yourself, is this organization truly an AI organization and are we thinking that
03:10 they’re going to be around long-term because they may be doing stuff that may look like AI, but they’re simply just washing over an API with an AI implementation that really doesn’t add a ton of new value. So, keep a look on this and we’ll keep monitoring it as we go. >> [music]