It's About To Get Real
We’re seeing the signs everywhere—AI is rapidly coming of age. A few weeks ago it was the whole Claude-bot/Molbot controversy. This week it’s a rash of cautionary warnings and senior team members resigning from different AI companies.
Probably the most disturbing thing I’ve read in a while is post from senior safety researcher of Anthropic, Mrinank Sharma:
https://x.com/rohanpaul_ai/status/2021449003414794330?s=20
Within the past few days xAI has also lost several team members who just recently revealed they're leaving to create a startup called Nuraline, which according to the company’s website, “Enables AI systems to self-improve in production.”
This is also a bit concerning as well because because tools like Nuraline will make it awfully tempting for AI-startups to justify unleashing their product onto the world before it's truly ready. It's now truly a race to see who gets there first.
We have lots of things to consider this week and yet I believe these rapid developements will be just a small glimpse of our the pace of our future.
Wherever you stand, philosophically, on the issue of AI it’s becoming crystal clear that it’s here to stay. This article by Matt Shumer has gone viral over the past twenty-four hours for a reason—although I don't agree with everything Matt says his article is filled with some profound wisdom.
https://x.com/mattshumer_/status/2021256989876109403?s=20
It's about to get real, folks. AI is evolving faster than ever before. No one really knows where this will lead humanity but it's never been more important to be paying attention.
All for now. Thanks so much for reading.
I was checking out the new AI Agents thing that Crypto.com released the other day. I got to the point where it was asking for a credit card and I decided not to go any further than that.
I also have reservations. I saw somewhere that I can download the OpenClaw AI agent. But then I read that it scans everything you have on your computer, all your files and links, so I changed my mind...
Yeah, these agents are a little concerning to me.
I'd be very cautious about handing over any information like that. I've already heard multiple horror stories coming out of the Molbot community. I just read yesterday that within 6-9 months AI-generated spam will be so bad that there won't be effective tools against it. The author predicted gmail and other email services will become unusable.
Oh great! That's not what I want to hear! It's surprising that they wouldn't be able to use AI tools to fight the AI generated spam... :) I still don't fully understand the breadth of what AI agents are and what they can do.
I wouldn’t be too concerned just yet. There’s a whole lot of doom and gloom out there right now.
If we move away from static model weights to systems that learn and adapt in real-time based on live feedback, we’re essentially losing the ability to audit the AI. It's becoming a moving target that improves faster than we can regulate it.
Exactly! It'll be impossible to catch up.
Interesting to see how these advancements will shape our daily lives and industries in the near future.
It will be!
Don’t worry, AI isn’t dangerous — and if it ever becomes a problem, we’ll just switch off the internet, cut the power, and go back to living in caves! 😄
Lol. It's frightening how many people still believe this will be possible.
My hair stands on end! What could these resignations mean: something new or disagreement with what has been done? I find the poem with which that letter ends very revealing. Revealing, inspiring, and thought-provoking. Hugs
It's an ongoing saga with the resignations. Originally the employees leaving xAI said they were doing so of their own choice to start new companies but later yesterday Musk made the announcement of an internal restructuring of the organization. I think AI is getting to the point where fewer humans are needed because the foundational work has already been done. Most AI platforms have the ability now to evolve on their own. Sharma's letter though is highly concerning. Elon has warned several times about corners being cut in terms of safety for the sake of companies wanting to be "first". It looks like his concerns were valid. Thank you Nancy!
Human stupidity will kill humanity and that's not too far...
The possibility of this ending badly definitely isn't 0%. It's going to be a wild decade.
100% agree with you
What he writes is what resonates with me a lot. Overall, the bigger pictures, I think Matt is correct, or close to corredt. Including the time it'll take to get there. Perhaps not 1 year, but 5 years seems a very very long time to me looking at how quickly the quality in AI is increasing.
I do wonder what parts you don't agree with. Please dont go into details :)