Twitxy Saved

Alistair Croll

Verified account This account is verified. Learn more

@acroll

Author (@leananalytics w/@byosko; @evilenough w/@emilyjaneross) & functional government (fxgov.com). I chaired that conference you went to that didn’t suck.
Montreal, Quebec acroll.substack.com Joined March 2007
18K Following 24K Followers 36K Posts

Latest posts 5

Pinned post

Alistair Croll Verified account @acroll
The asymmetry of the modern age has been: the institution has machines and bureaucracy; the citizen has a phone and their patience.
What happens when the citizens can build machines of their own in a matter of hours?
alistaircroll.com/updates/the-ma…
www.alistaircroll.com The machine fights back The machinery of government grinds at its own pace, and citizens have been forced to endure it. But cheap, abundant software means residents now have their own machines—and the state is not ready for...
1 replies 1 reposts 9 likes 1.2K views View on X
Alistair Croll Verified account @acroll
Marketing has a local maxima problem.

This thread is gold, and speaks to a complete failure of the imagination: marketing is about creating attention and turning it into behavioural change.

An algorithmically mediated world drags us inexorably towards sameness and unremarkability. This isn’t just a hot take; it’s actually math (see laurenleek.substack.com/p/temperature-…).

Marketing is the resistance. Done right, it is exploration over exploitation.
laurenleek.substack.com Temperature Zero for Culture: Why Everything Is Starting to Look the Same What 640 London shopping streets, 4,000 Billboard hits, a million films and synthetic AI personas tell us about why everything is flattening into sameness and how to regulate the algorithms behind it.
Rory Sutherland Verified account @rorysutherland Nov 3
This is 100% bang on the money. By the way, one of the fundamental assumptions of mainstream economics is that people already know exactly what they want and can define it in advance.
View quoted post on X
Alistair Croll Verified account @acroll
Ramez doesn’t post enough, and when he does, it’s always an excellent read.
Ramez Naam Verified account @ramez Oct 19
🧵New post in my AI series: Can a million parallel agents create superintelligence? TLDR: More agents buy speed, but intelligence gains diminish fast, and much faster than RSI models. Thanks @noahpinion for the encouragement! 1/25
View quoted post on X
2 replies 0 reposts 13 likes 921 views View on X
Alistair Croll Verified account @acroll
Last year I said SaaS had a 24 month shelf life. I was not thinking about the dated law around Clean Room reverse engineering.

If you’ve seen Halt and Catch fire, that was a clean room: one person documents a system, hands the spec to a second person in front of a lawyer, and the second person implements the spec.

But now that we can hand code to an AI to decompile, write a spec, and implement that spec, Clean Room approaches are likely to face new legal scrutiny. That’s not a thing the patent and copyright offices anticipated.
Eric S. Raymond Verified account @esrtweet Oct 6
This is the doom I predicted a few days ago, coming for Photoshop. A clean-room open-source reimplementation.

No prizes for guessing that they decompiled Photoshop to source code, processed that to some kind of non-code specification language, then fed the spec to an LLM with an instruction to generate Rust.

Adobe just got nuked. And closed source is dead, dead, dead.

github.com/storytold/phot…
View quoted post on X
3 replies 2 reposts 31 likes 3.6K views View on X
Alistair Croll Verified account @acroll
This is so true. Legitimacy and reach are the new currencies. We learned a lot from releasing @evilenough and broke many of the industry’s rules in doing so. @emilyross knows of what she speaks.
Emily Ross Verified account @emilyjaneross Aug 16
AI Prediction; We're going to see a major reversal in publishing trends.

The cost of creating plausible (superficially readible) books is rapidly approaching zero, thanks to generative AI. And according to a recent Business Insider report we're already seing an explosion of AI generated books on Amazon.

The friction of manufacturing and publishing has already been lessened by self-publishing.

Gen AI + self publishing + Amazon print on demand = near infinite books.

The supply of books will become (effectively) infinite. Though many of those books will be iterative and mediocre.

So what is the new limiting factor? It's the ability to persuade a reader that an author's work is worth reading. That's going to be infinitely harder to do, especially for unknown authors. Self publishing will be far less viable for authors without pre-made audiences.

Publishing stops being about book manufacture (though bricks and morter distribution, especially into libraries, universities and high street books stores remains somewhat gated)

Publishing now starts being about curation and trust. In short; The publisher’s value shifts from “we can *publish* this” to “we *chose* to publish this.” The gate becomes the product.
View quoted post on X