
Founder of @bromleymindfulness and @meditationclass
London, England
Joined October 2007
- Tweets237
- Following259
- Followers279
- Likes18.9K
Robert Mitchell retweeted
A slide from @AnthropicAI's IPO roadshow. Hits hard and is very true.
AI is soooo much better than how most people are using it.
Replying to @dani_avila7 @mustafa_2vec
I now specify the technical part of my role primarily as an "Agent Fleet Controller".
A good starting place is to get Fable 5.2 to read all of your projects (specify the level, possibly from git) and create a graph diagram. Entities, edges and attributes. Then push the synergies.
1
67
Replying to @IntuitMachine
What a ridiculous conclusion.
Flipping between conscious and unconscious states 🤣
Obviously, doh.. the 20ms of sound is a footprint that is recognised by the brain because it hasn’t deviated enough in pitch tone and volume from the sound when played forward…
10
This!
Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less "two groups of people speaking past each other" and more a sharp funnel.
- Napkin math somewhere around 6B people (~75% of the population) have barely come in contact with LLMs at all.
- Around 1-2B (~20%) are casual and infrequent users of free-tier ChatGPT-like products. This group sees derpy chatbots and treats them a bit like a better Google search, a writing aid, or etc. Maybe an agent tries to book you a flight. I have non-tech friends who (reasonably, imo) say they have not much use for it at all. Even many professionals outside of math&code are in this tier. For example, execs and many other functions spend a lot of their time talking to other people, so while they understand what is happening intellectually, it is still second-hand and a bit abstract.
- Now we get to professional use of frontier-grade LLMs in math&code. Somewhere around 20M people (0.2%) see first-hand that large, complex projects that used to take them weeks/months can now be completed by agents with a prompt. Building apps, copying apps, translating apps, decompiling apps from binaries... This has all happened very quickly and recently - less than 1 year ago, I was writing code manually by hand, typing memorized computer code commands into a code editor character by character, occasionally pressing Tab to autocomplete a little chunk of code.
- And finally we get to the ~5,000 people (~0.00006%) with access to frontier-grade systems internally. The external world has seen the preview. It looks like swarms of thousands of agents collaborating over weeks on software mega projects: minting zero days, running cyber attacks and defenses at machine speeds, discovering new science, advancing the frontier of mathematics. Things that would have taken top professionals in the industry years of work. Meanwhile, human review and comprehension are starting to fall behind. For example, people are still involved in the "archeology" of the OpenAI-HF incident from many months ago. Mathematicians may be poring over the 722 manuscripts on frontier mathematics for a while.
The funnel is driven by a combination of factors:
1. The impact scales with ambition, problem size, and horizon. A question with a paragraph answer barely stresses the system. You need a reservoir of big, difficult problems that you really care about. This aspect drives the consumer / professional dimension of the funnel.
2. The jaggedness of the system (which I have written about a lot separately). Capability peaks in domains that are digital, verifiable and economically valuable. This is because LLM capability emerges from reinforcement learning on verifiable rewards on a curated environment mixture driven by revenue potential. This aspect primarily drives the area (e.g. math&code) dimension of the funnel.
3. Access. Free-tier, paid-tier, internal.
So this is the weirdness of the moment. The general public has mostly not interacted with these systems. When they have, it looks like a derpy chatbot. The majority of professionals still see only a modest uplift. And a small sliver of professionals are experiencing the vertigo of the curve going vertical. And it is all happening at the same time.
2
11
Robert Mitchell retweeted
If you have a problem with what is happening to our society, manifested by a small but loud number of activists… yet you say nothing! Do not complain when all common sense, logic, respect, civility, debate, small businesses, free speech & womens rights are gone!
Replying to @bindureddy
Sounds wonderful but no link and no sign of”Abacus AI Bot” on your site logged in or out of my account 🤔
1
1
594
Replying to @Voxyz_ai
It’s far simpler than that. Instruct Fable to create an efficiency report and a process to manage your token use. 1. Flag and remove redundancies and duplications. 2. Write scripts.3. Delegate to other models.4. Use skills more.
2
89
Replying to @unclebobmartin
It is marking its own homework. Incorrect assumptions may well live in its context.
I have agents I call Validators which do QA and validation work within automated processes and I trigger them by hand when needed too.
Then get the agent to review the validation report.
84
Replying to @MatthewBerman @bot
I have to give it a try - so much chatter about Grok Bot and I currently run an agent fleet on Nanoclaw so it will be an interesting comparison.
31
The hysteria about AI #watermarking is hilarious. All these people that want to pass off AI text as theirs.
1. If it's your writing why would you care?
2. If it's AI generated what's wrong with stating that?
Surely text/copy stands on its own irrespective of the source.
1
34
Replying to @svpino
M$ don't need to do anything like watermarking because they simply give governments a back door. Remember, @anthropic told the Pentagon they couldn't have access to Claude for military use. Why would you care about watermarking? Do you want to pass off AI text as yours?
1
70
Replying to @HamelHusain
Instruct it to use “Simplified Technical English” and over time tweak it’s claude.md to respond the way you want. I have a specific agent I have named “The Narrator” that I use to refine responses. From that, you can construct voices for different tasks/roles.
1
276
Replying to @AndrewCurran_ @VoidNulled
This will lead to some hilarious outliers. We will be back in the days of “delve”.
1
193
The improvement of my performance by using AI is directly proportional to the learning curve of the technology or skill I'm using it for.
That isn't how complex it is.
It's how unintuitive it is.
6
26
Replying to @jon_stokes
Clear the context and if that doesn’t work, adjust the effort down and separate what you want done into separate tasks that are test first. Which is what you should probably be doing anyway 😉
1,591
Three months ago the same guy was telling the WEF about “AI garbage workslop”
nitter.gq/FundamentEdge/status/2…
To contrast, this is what Ken said just 3 months ago at WEF:
“There was a recent Harvard paper on this they called it AI work slop — that it looks good but if you sort of peel back the onion, the substance isn't there. I was with one of my colleagues who runs our commodities business, and they handed a report on that we were generating with an AI engine. Doesn't matter what the topic was, the first few sentences — wow, that's really insightful — and then you go down below that and it's all garbage.”
46




