Jim Roslund
36 posts

Jim Roslund
@RoslundJim
I don’t just build companies, I try to build something that matters.
Sweden Katılım Nisan 2020
23 Takip Edilen8 Takipçiler

Ex-Goldman Sachs Insider: Why Bitcoin Could Hit $140,000 Soon beincrypto.com/bitcoin-140k-g…
English

@alliekmiller When does AI stop being a productivity multiplier and start behaving like economic infrastructure with second order effects on capital allocation, org design, and who actually captures value?
English

There are actually two ways to use AI for research.
The first is obvious. The second is a superpower.
The first is to speed up what you were already going to do. Duh. Researching competitors. Looking up people before meetings. Background checks on companies you are evaluating. These save time but the scope stays the same.
The second is more interesting. AI also lets you research things you would have never spent hours on manually. Cross-industry pattern matching, pulling insights from fields you would never think to look at, synthesizing all enterprise docs from another unrelated project.
(I don’t know if anyone has seen the movie Working Girl but it inspired this method.)
Try a prompt like:
“I am a [your role - marketing manager] at a [your industry - pharma company] preparing for [your task - a new product launch] targeting [your audience - physicians]. Pull insights from at least 5 industries outside of [your industry - pharma] on how they successfully [your goal - launched products to highly educated, skeptical professional audiences]. Build a report on the top patterns and then create an abstraction layer that translates each one into something I can operationalize in my context. Explain why you believe it will work and provide open questions or risks and an action plan I can use to fill in the blanks or mitigate risks.“
This is where AI research gets powerful. And actually, this is where AI in general gets powerful. Not just doing things faster. But doing DIFFERENT things.
I spent years working across industries at AWS and IBM. One of the most valuable things was connecting rooms, eg sitting in a conversation with a media company and then recognizing the same problem at a manufacturing company later that week and using what I learned in Room A for Room B.
The best practices transferred from one industry to another. The patterns connected. Patterns the other room would have never been exposed to. But it required being in both rooms.
AI can now be in every room. It can surface latent connections across domains that would take a human consultant years of exposure to recognize. The constraint on cross-industry learning used to be access, now it’s just knowing to ask.
English

@fundstrat Where do you see the bigger mispricing right now: ETH vs BTC, or ETH vs AI equities trading at 20–30x sales?
English

@fundstrat Do you see this as the start of a sustained correlation regime between equities and crypto or is crypto still just front-running liquidity before equities catch up?
English

Check out my latest article: How do you summarize 2025 from an AI and IT consulting perspective? And how do I view 2026? linkedin.com/pulse/how-do-y… via @LinkedIn
English

Still bullish on #Ethereum. Tom Lee isn’t wrong, liquidity + network effects point up. PoS + on-chain activity = real infrastructure. I’m long $ETH. What’s your signal? @fundstrat
#ETH #AI #Web3

English

Check out my latest article: The illusion of progress. The next AI superpower won’t be a country, it will be collaboration. linkedin.com/pulse/illusion… via @LinkedIn
English

