A 95% AI project failure rate is actually the sign of a healthy experimentation phase, not evidence that the technology doesn't work.
Arvind Jain reframes the widely-cited MIT statistic that 95% of enterprise AI projects fail as expected and desirable given the pace of experimentation. ✦ AI generated
Arvind Jain · BG2 Pod · 2025-12-23 · original ↗
starts at this moment · 2:35
“What's the reality? Bridge that gap for us. Lay it out as you see it, a view from the top.”
You hear these 95% of projects fail, but like you know, like that's that's that's actually what you want. Like you you like when you are actually experimenting with new technology, if if if all of all of your projects are failing, that means you didn't just not trying enough, you know, at the moment.
verbatim transcript · starts at 2:35
2:35is you you hear these 95% of projects fail, but like you know, like that's that's that's actually what you want. Like you you like when you are actually experimenting with new technology, if if if all of all of your projects are failing, that means you didn't just not trying enough, you know, at the moment. So, so I think when when I read the study like it was not a surprise for me.
2:53Um you know, we're going to actually see hopefully like you know, similar stats next year, too uh because you want everybody in the industry to be to be really eager and experiment and actually figure out like you know, how to mix how how how to actually make you know, get you know, get benefits from this technology. This would make you guys by default the 5% of AI that is working.
3:12>> [laughter] >> Which is uh one in 20. Maybe maybe you go to the 5%. What is what is a use case that is working? And not just working like it's like saving me time, but like it's working and it's transforming my company. Something that you can take to the bank to to the CFO while while the CFO will not listen, but the legal won't shut it down. All right, well, I mean,
3:32look, we're seeing a lot of use cases that are working. Uh it's just that you know, you you just have to it's not just you can just unleash the agent and it just works. Uh it's an engineering art. Like if you're going to have a company that's going to be really differentiated like like my company or your company or anyone's company and you want to beat the competition,
3:51you can't just like you know, quickly put something together and think that you know, the your competition is not going to do the same thing. So, that's going to be you know, something that needs evaluations. It needs something that you know, you're going to productionize. It's going to take effort. You need a great team around it. But we're seeing a lot of them. Like I'll give you some examples.
4:07Um Royal Bank of Canada uh built agents with us that basically take as soon as an earnings report comes out, so equity research analyst, their job is to put together these, you know, reports that say like you know, this is a buy, this is you know, hold and so on. Um the agent goes, gets the earnings report, gets all the previous earnings reports, gets all the competitors' earnings
4:27reports, gets everything that's going on in the market, does the full analysis, the news, everything, puts it all together and it can get the equity report out in 15 minutes from the earnings call. Industry standard is 2 hours. Of course, it's going to get commoditized and others are going to do that as well, but uh that's actually really important use case that we're seeing in finance. Um so, that's like finance example in