33 Comments
User's avatar
Jen H's avatar

Thank you for this. For those of us who are still in the shallow end, learning about AI, it's challenging to tease apart the myriad legitimate concerns from those we probably shouldn't be losing sleep over (at least not yet).

Expand full comment
Nate's avatar

glad it's helpful!

Expand full comment
Jurgen Appelo's avatar

Agreed. One consistent theme throughout humanity's history is that the doomsayers are always proven wrong.

Expand full comment
Irina Kilimnik's avatar

Thank you, Nate, every post is real, interesting and very educational.... My ChatGPT calls you a brilliant guy lol. Far from being any kind of AI expert myself, I've been using it constantly across different subjects - and I keep coming back to this: the real problems are still all human. So far, we've shown no ability to create or train anything beyond our own limitations.... It's a tool and a toy, for now... until something emerges.

Expand full comment
Nate's avatar

haha thank you! I think “the real problems are still human” could be a t-shirt honestly. I’d wear one. It’s just so true and not said enough

Expand full comment
Obie's avatar

Would love to hear your take on “AI 2027” and/or what’s laid out in the video here:

https://youtu.be/5KVDDfAkRgc

Expand full comment
Nate's avatar

I mean this kind of is my take on AI 2027 lol—I literally mention it in the piece

Expand full comment
Obie's avatar

I rewatched your video, I have no idea how I missed that other than I was multi-tasking at the beginning, my apologies. I appreciate the reply, as well as your take. Completely agree that there are outcomes that are not as cynical but just as possible.

Expand full comment
Angie's avatar

Thank you for a balanced perspective on these hot buttons regarding AI. I have followed you for awhile and always appreciate your view and educational contribution to how we can effectively work with AI.

Expand full comment
Nate's avatar

Glad it was helpful, Angie!

Expand full comment
Tim McAllister's avatar

Spot on, Nate! The existential risk debates are a sideshow compared to the chaos we can already see coming from unauthenticated, AI-generated content with malicious intent to defraud or virally disseminate misinformation. Coalition for Content Provenance and Authenticity (#C2PA) and cryptographic provenance aren’t just technical nice-to-haves—they need to be mandated globally if we want any hope of containing the next wave of viral disinformation. Thanks for cutting through the noise and focusing on what actually matters.

Expand full comment
Nate's avatar

It would be really helpful to have some kind of content provenance (“this is a sparkling AI white paper, real AI white papers come from the domain of…” jokes write themselves)

Expand full comment
Tim McAllister's avatar

Totally—think of C2PA as the “nutrition label” for digital content, except it’s cryptographically sealed.

If a white paper (or meme, or deep-fake video) lands in your feed without a signed provenance manifest, your browser should throw the same warning it does for an invalid TLS cert: “Source unknown—proceed at your own risk.”

Once publishers start embedding those manifests and auditors can trace the chain (sigstore/Otterize, etc.), the sparkle will sort itself: real diamonds show up with a certificate; cubic-zirconia AI and spam doesn’t. Until we get that layer baked into every content pipeline, it’s caveat lector.

Expand full comment
Liz Homer's avatar

I wish I knew more about what you mean by LOM. I don’t happen to work in your field.

Expand full comment
Nate's avatar

Do you mean LLM? I wrote a guide for beginners here: https://natesnewsletter.substack.com/p/the-a-to-z-ai-literacy-guide-2025?r=1z4sm5

Expand full comment
Rob G.'s avatar

Way more concerned about humans/governments using it to hurt other humans than AI turning on us. As for ASI, I forget if it was Hinton or Harrari (maybe neither) that compared humans to ants. We may see an ant on the sidewalk and not step on it purposely, but that ant is also insignificant. We may some day be ants to ASI. Like how we share distant ancestry with ants, we don’t look at them that way. Humans similarly will be AIs origin story, but we won’t matter.

Expand full comment
Nate's avatar

Yes, Hinton has said something like that. We’ll have to see how it goes—what a wild time to be alive

Expand full comment
Joseph Ibrahim's avatar

Nate! This is another incredible fireside chat. I enjoyed how you turned a talk about pdoom into a talk about advancing education and handling deep fakes. I'm a parent, and I find increasing value in these lessons. This is applicable while I type this comment. You are crushing this substack with tangible insights.

Expand full comment
Nate's avatar

Thanks Joseph! Glad you found it helpful. Being a parent is definitely a tricky job these days

Expand full comment
Philip Wilkinson's avatar

Excellent post. We have those much shorter-term risks and challenges to focus on while we messily build this out. In the history of AI, we always invent a new technology or approach, it rises rapidly, then it plateau's until the next innovation. Perhaps LLM's will at some point hit that wall too.

Expand full comment
Nate's avatar

thanks Philip!

Expand full comment
Kathy  Utley's avatar

Love the idea of educating seniors and helping young people

Expand full comment
Nate's avatar

it certainly seems better than panicking lol

Expand full comment
Jason's avatar

I am thoroughly convinced AI will NOT take over the world. I tried getting AI to create illustrations for a children's book. It made an illustration in which I only wanted the color of the shoes on the pig changed. I wanted everything else to stay the same. But the updated the illustration changed the background, the apron on the hen, the rain hat on the duck, and on and on. I literally told the AI to keep EVERYTHING THE SAME and change ONLY the shoes. It couldn't do it.

Expand full comment
Nate's avatar

haha it’s the new “lipstick on a pig” joke but for AI

Expand full comment
Mike Gaffney's avatar

Unfettered capitalism and lacking a strong social network isn’t going to be good for a majority of Americans. Now with the Democrats groaning about another social Democrat running for office I throw up my hands.

Expand full comment
A.J. Cave's avatar

Fully agree! I recall a 9-minute mini-lecture by quantum physicist Michio Kaku in the bonus section of the Total Recall (2012) remake. Most of our fear of emerging technologies [AI in this case] comes from science fiction not science.

Worth a watch:

https://youtu.be/rRzfZ2vZwXg?si=unUyMTKkqQYUvp3S

Expand full comment
Nate's avatar

Yes, sci-fi isn’t doing us any favors here. Thanks for the link!

Expand full comment
CJ Pulido's avatar

I love it! I’m still advocating for the idea of simply preparing for the future where things will be okay. You can’t prepare for an existential future. More emphasis needs to be placed on the current education system. It’s too slow to adapt.

The skill of being adaptable and agile is crucial. Just like how it’s impossible for a waterfall project to compete with an agile project in today’s tech world.

Expand full comment
Nate's avatar

exactly, you can’t prepare for an existential future! Well put

Expand full comment
Charles's avatar

Nice! Part 3 of the letter series. I would love to know if there was an equivalent equivalent with the advent of cars or internet, heck, industrialization and factory automation. I bet there are versions of this which would be cool to see put forth by people settling evolutionary alarmists. Thanks for calming the rabble rousing Nate!

Expand full comment
Nate's avatar

there absolutely was, but I need to find the receipts for that

Expand full comment