Today, we’re launching Clone.
An agent harness that predicts your next prompt.
- Clone Mode keeps prompting your AI agents and never stops till done
- 93.9% prompt prediction accuracy, outperforming Astra and Opus
- You can use your own ChatGPT or Claude subscription
Want to push agent-maxxing to the limit?
clone.is/you
I agree with the No AI Slop manifesto EXCEPT the fourth point.
It says AI “cannot replace human judgement, it can only amplify our judgement (or lack thereof).”
AI already outpaces us on many tasks. It is already asymmetric, and on top of that, the gap is growing exponentially. Now we have to make decisions one at a time. But Clone could apply my judgement to a billion or even a trillion decisions at once.
Dogs evolved with us for thousands of years. Even the muscle that lifts the inner brow is more developed in dogs than in wolves, because a more expressive face may have helped dogs win the affection of the people who fed and sheltered them.
We, too, will change as we coevolve with AI! The AI Clone marks the beginning of an ontological shift in what it means to be a person. Those who don’t delegate their judgement and put it to work in parallel are already under evolutionary pressure.
Korea’s top tax-exam scorer admitted to cheating with AI. The country is in an uproar. Its college entrance exam is days away.
AI was already capable. The public takes notice when it threatens the system.
We see referrals as favoritism. We equate exams with fairness. Our civil-service exam tradition dates back to 958.
Can we adapt when AI challenges the very system we trust to decide who deserves to get ahead?
Our YC application video.
Hollywood’s best are already using Shoty.
We’re DoubleZero. We’re building a beast of an AI video model—not another API wrapper.
@garrytan@ycombinator, good luck saying no.
Feels like everyone is having harder time with online marketing lately.
So we made Clone keycap keychains!
Want to make a limited edition soon that actually turns on "Clone mode" when you press it.
We’ve put a lot into Clone.
So happy to finally share the beta!
Astra, Fable & more are free to try. Drop your use case on the original post for a $1 first month of Plus (then $20/mo).
Introducing Clone, the Personal AGI that predicts your next prompt.
Press Tab twice, and your Clone keeps prompting your agents.
Clone-in-the-Loop just killed Human-in-the-Loop.
Available now (Beta): clone.is/you
Comment your use case, and we’ll DM you a $1 first-month coupon for Clone Plus ($20/mo).
Introducing Clone, the Personal AGI that predicts your next prompt.
Press Tab twice, and your Clone keeps prompting your agents.
Clone-in-the-Loop just killed Human-in-the-Loop.
Available now (Beta): clone.is/you
Comment your use case, and we’ll DM you a $1 first-month coupon for Clone Plus ($20/mo).
Your AI voice sounds human. So why can't it say your product's name?
A great AI voice reads "Porsche Taycan" as TAY-can. Porsche says TIE-kahn. It guessed from the spelling, and nobody caught it, because nobody listens to line 1,200.
Today we're launching Onepin: the production step after text-to-speech. It checks every line of voiceover before it ships using the voices you already work with.
Onepin can:
➤ Check people's and product names against a 4-million-word pronunciation dictionary
➤ Spell out prices and dates before the voice speaks
➤ Score every line of audio for naturalness, clarity and word accuracy
➤ Fix the one wrong word in the same voice, without re-rendering the take
Works with your voice subscription on @ElevenLabs, @OpenAI, @Google and 30+ more. No phonetic spellings to type. No re-rolls. No switching providers.
Free to start, no credit card required. Hear the before and after in the thread ⬇️
Students sometimes ask me if it still makes sense, in this accelerating age, to pursue a PhD in AI.
Perhaps counterintuitively, I think it's a great time to do so.
I wrote up some thoughts on this here: web.mit.edu/phillipi/www/wri…
I'm getting used to handing more and more purchasing decisions over to AI. Recently, I've used GPT 6 Astra to:
1. Find a vendor to 3D-print a laptop lid stopper for Clone.
2. Find a dermatology clinic near our office.
3. Recommend literature to read: poetry, novels, and so on.
I was pretty happy with the reading recommendations. I gave it plenty of context about the writers and books I already loved, and the suggestions felt right.
The first two were less satisfying.
What interests me is how easily all that dissatisfaction gets filed under “the AI wasn't good enough.”
But a disappointing outcome can have very different causes:
1. I may have given it the wrong criteria. For a dermatology clinic, perhaps I should have asked it to weigh reviews, pricing, and other factors together, rather than prioritize proximity to the office.
2. The relevant information may simply not be available online. When I asked for nearby 3D printing vendors, it found businesses near here. Only after placing the order did I learn that their actual production facilities were all in quite far from here.
3. Its answers may have been skewed by the flood of AEO/GEO-optimized content.
4. The choice itself may have been sound. Sometimes you simply happen to get a disappointing product or service.
If I felt I had made the decision myself, I might have been more satisfied with exactly the same outcome, for two reasons.
First, visibility. Having explored the range of available options, I'd understand why this was the best choice, or at least a reasonable second-best.
Second, rationalization. Even if it wasn't either of those, it's easier to convince myself I made the right choice when the choice was mine.
Will this feeling fade as we get used to making more decisions through AI? Will interfaces get better at showing us the options and trade-offs? Or will they get better at making us feel as though we made the decision ourselves?
There's another problem: AI can’t get feedback.
Unless we make a deliberate effort, our satisfaction or dissatisfaction never becomes part of the context the AI can learn from. Whether the recommendation works out or not, we tend to go quiet once we have an answer. Why spend more tokens on a conversation that already feels finished?
With programming, the feedback loop runs largely inside the computer. Errors, test results, and other signals can keep flowing back to Codex.
For decisions like these, the loop breaks at the point where I have to go out into the world and experience the result myself. The AI gives me a recommendation. I live with the outcome. Unless I come back and tell it what happened, the loop stays open.
I'm going to try keeping detailed reflections on these decisions and feeding them back. But I'm still unsure whether that's an efficient or useful habit, or whether I'll be able to tell if it actually leads to better decisions over time.
1. This is the most thought-provoking piece I’ve read in a while. The author’s incisive analysis gets to the heart of what intelligence is.
2. How I came across it makes it all the more interesting: through a carousel posted by a Korean Instagram account, rather than on X itself. With Instagram’s format built so firmly around images, I was used to encountering lighter, more mainstream content there. I was surprised to find a real audience for a post like this.
3. Even on an internet that seems to be growing flatter, then, there is still plenty of room for translating value and for arbitrage. Once, there was plenty of value to be captured simply by translating Steve Jobs’s Stanford commencement address into Korean. Today, automatic translations let people watch videos of well-known figures directly, even when English isn’t their first language.
4. Translation is now needed for discovery, rather than comprehension. With the editorial discernment to spot a Chinese-language post on X and turn it into an Instagram carousel, you can still cultivate a real audience. A line from Anna Tsing’s The Mushroom at the End of the World, which I’m currently reading, has stayed with me:
“Translations across sites of difference are capitalism: they make it possible for investors to accumulate wealth.” (p. 62)
This may not be the most faithful use of a passage from a book that explores the possibilities of survival in pericapitalist worlds. Still, her words have become entwined with this post in my mind, and I wanted to put that connection into words.
1. Outsourcing execution
2. Accelerating execution
With these two shifts brought about by agents, I feel we are moving toward a way of working that makes it harder and harder to stay absorbed in a single task for long. Losing that sense of immersion is frustrating in itself, and it also makes us less effective at our work.
To get the most out of agents, you inevitably end up running many sessions in parallel. Yet, as Garry pointed out in a recent YC talk, our working memory has room for only about the “magic number seven.” Juggling different kinds of work, each complex and laden with context, is an ordeal. I feel like a seven-digit computer suffering a cache miss on every read.
Most of my “bio tokens” now go toward reading and making sense of what I have read, because agents handle most of the writing and execution. Reading, meanwhile, demands an “Extra High” level of mental alertness and focus, so the number of hours I can devote to it each day is more limited than I would expect. Lately, I cannot shake the feeling that, despite having the ability to put agents whose knowledge far exceeds my own to work in parallel, my biological limits keep me from making full use of them.
How can I create the conditions for deep, sustained focus again? I have been trying what I can at a personal level. I tend to eat smaller meals, for instance, because I find it hard to concentrate unless I am a little hungry. But I wonder how long I can keep a clear enough view of everything to manage what feels like running an entire city. And as we run a race toward a destination already set, I wonder whether we really have the luxury of choosing between speed and the scenery.