The Law of Displacement & why it will ruin your AI Agent big product idea 🤖
AI is both ruining and disrupting SaaS right now. And it's a problem. Because these lazy "get-quick-rich" AI pseudotools can ruin the whole industry's reputation - and make AI-powered solutions get a bad rep with users (I see you rolling your eyes every time I type "AI" already) - even for the legitimate ones that can change the way we use software. Like LAMs - Large Action Models that power Generative Process Automation - which will collectively shave years off how long it takes us to complete certain workflows in complex tools like CRMs.
MVP stands for Minimum Value Product now?
Kas Szatylowicz, who used to be the Head of Growth for several of AI tools like V7 and Lakera since the 'early days' of AI (before chatGPT became widely available - she also happens to be for hire now!) - wrote a great post about these "AI tools" - which are nothing but lame GPT wrappers with dubious prompts that don't reflect real user needs.
Because whenever there's a hype - there will be hype-artists that will ride the wave just to make a quick buck, without any consideration for the consequences for the industry as a whole. Enter the era of "Minumum Value Products":
As you can see, even from the AI graphic. But some AI solutions are going to be huge. Take AI agents - Shyvee Shi, product lead at Microsoft and ex-product at LinkedIn, recently penned a great interview featuring Bella Liu, Co-Founder and CEO of Orby AI - on how Generative Process Automation is causing a paradigm shift in how we're gonna be using software, and how much productivity we can unlock from it. Finally:
The problem is, some agents are too early.
I've recently demoed a few AI agent tools that promised the world and...just like Elon's robots,massively disappointing.The concept is there - but the state of the technology often doesn't really allow for execution at a level that wouldn't be disappointing to the point where...there's no point in using the technology. Take AI SDRs. Take them, and burn them with fire.
Recommended by LinkedIn
But as Jason Lemkin wrote himself, "yet." So it's good to be early. But also not too early...to be confused with these "Minimum Value" trash Products that Kas wrote about - if you know what I mean?
But there's another problem with this disruption...who owns the platform?
Before you get too excited about disrupting a certain process that annoys you - let me rain on your parade - have you heard of the "Law of Displacement Speed"? No?
Basically, ask yourself the question - who owns the platform that operates the process you want to disrupt?
The platform owners are the most likely to a win at the game - because they own the real estate and the whole ecosystem where the AI agents can operate.
So why wouldn't they drive your little solution out of business? But also, if you're lucky - they will buy your solution. That is - if you're fast enough to build a critical mass of innovation, it's just easier for them to buy you than replicate what you did...but not too fast to be deemed useless.
Now, are we building AI agents at Userpilot? Maybe :)
Hope this was inspiring! See you next week!
Emilia from Userpilot
B2B Freelance HR Content Marketing Writer specializing in employee engagement, talent management, recruitment, and L&D.
4wI have yet to see good AI use cases that justify the $100+ billion investments. It's been two years since the release of ChatGPT and all we have are a bunch of AI "agents" and a 10% better ChatGPT. This is slowly becoming another "crypto will change the world" situation
Ecosystem-Led Growth Marketing @ Demandi.co. 4X girl dad. Army veteran. Acquired 2 HVAC businesses and looking for a 3rd.
4wI don’t know if I agree with the sentiment here at the end of the post. Fast movers are the one who prove out use cases and bring to light the next few steps up the hill. For a majority of those that are unwilling to devote time and effort to doing the mental gymnastics around the hyped tech,this IS the only way they begin to see what’s possible. Or to see the promise. The validation of the demand as it ebbs and flows with what seems to be headlines from quarter to quarter. I’d argue they make it easier to get funding because they create the context for understanding actual potential. Many generative use cases died immediately after launch, and many more are limping along. Those they can survive, and also should, likely will. I mean the goal is to sell hotdogs to hungry people, not those with full bellies. People who start anything for which there is low demand are the large portion of the 95% of startups that don’t make it. Hype creates an opportunity for fast movers to harvest that demand by solving a problem. It’s how markets work at their core and why they continue to work for even proven technology. The thing offered is better than the alternative or the opportunity cost of doing nothing. I say let em innovate!
Your Girl's Guide to Rebuilding Your Personal Brand as a Migrant Woman
1moThis was an interesting read!
Product Management Leader and Advisor | SaaS | B2B | Sales Enablement
1moTerrific article. It's not surprising to see the trend of overpromising on new technology. Don't blame the creators of these products but it is definitely buyer-beware . It might be hard for businesses to resist the promises when they think the product can fill an immediate need. They are not only placing a bet on the current product features but on the developer. Can the chosen developer keep up with and outrun competitors as the market evolves? Investing in a dead end is tough on company success and decision maker careers.
Co-founder of Endurance Learning | Building #Soapbox to help improve every presentation!
1moThis is my read of the day for sure. I'm going to leave with the message that word of mouth and reputation will matter most in 2025.