iiinova — where your digital self keeps evolving

Vitalik Buterin @vitalik.ca · 04/10/2026

Doing a bit of a self-experiment. Goal: use my personal health and travel data to provide personalized diet and exercise recommendations for me, using frontier models but in a way that avoids leaking to them any private… firefly.social/post/ff-af61208b6977…

4133

Vitalik Buterin @vitalik.ca · 26/09/2026

Testing out some of the mobile offline local knowledge apps that people have been trying to build (see here poidh.xyz/mainnet/bounty/31 ) Definitely getting much better than the one I tried to build myself 2… firefly.social/post/ff-81a8d2725e6c…

2140

Vitalik Buterin @vitalik.ca · 26/09/2026

Elliptic curves are not elliptical and are not curvy

1101

Vitalik Buterin @vitalik.ca · 26/09/2026

Reminder: you can now sync an ethereum node within half a day and with aggressive settings the space it takes up on disk can be under half a terabyte. EIP-4444 and hard work by client teams on optimizing snap sync has… firefly.social/post/ff-c31d98ea16d4…

1190

Vitalik Buterin @vitalik.ca · 18/09/2026

One small example of the core d/acc thesis (improving passive defensive technology improves safety and at the same time improves freedom and most other properties of social organization that we care about) is soundproof… firefly.social/post/ff-dcc22fed8283…

0102

Vitalik Buterin @vitalik.ca · 18/09/2026

Screenshots of the Philip Morris website. (reminder: they're a leading cigarette company) It's a good calibration point for how good modern marketing is at dressing up pretty much any corporate (or, for that matter, government) behavior and making it sound responsible and safe.

1192

Vitalik Buterin @vitalik.ca · 17/09/2026

Qwen 3.8 flash is truly impressive, and llama.cpp has been rapidly getting better and better at processing it columns are: pre-existing prompt, new prompt, generated, input tok/s, output tok/s This is on my laptop (strix… firefly.social/post/ff-67231e4cc912…

0131

Vitalik Buterin @vitalik.ca · 17/09/2026

AI on how Mozi would feel about modern effective altruism - the similarities and the differences (especially with the AI-risk-focused variety)

120

Vitalik Buterin @vitalik.ca · 17/09/2026

Since many people are learning about effective altruism, it's worth learning about Mozi, the ancient Chinese philosopher who was surprisingly close to being a proto-effective-altruist (particularly the… firefly.social/post/ff-11a0871a0db8…

4121

Vitalik Buterin @vitalik.ca · 08/09/2026

Fun though in-retrospect obvious fact: tort and торт share an etymology.

250

Vitalik Buterin @vitalik.ca · 02/09/2026

This happened to me 5/5 times with not-logged-in ChatGPT.

4130

Vitalik Buterin @vitalik.ca · 10/08/2026

I updated my 2023 roadmap diagram to overlay where the items that were there sit in the current Strawmap ( strawmap.org ). In general, a lot of overlap, but: * Some things got reshuffled in order (eg. quantum… firefly.social/post/ff-20089e3c5efe…

072

Vitalik Buterin @vitalik.ca · 07/08/2026

Minimax H3 is the first *open* video model I've seen that outperforms HunyuanVideo 1.5 - which is actually an impressive accomplishment for @TencentHunyuan to have held that throne for so long. For a while now I've been… firefly.social/post/ff-a775e498c8ed…

0120

Vitalik Buterin @vitalik.ca · 20/07/2026

But AGI is not the same as ASI (artificial superintelligence). To see the difference, let us zoom into what things might look like one year before the singularity. We also need to add a third line to the chart: the centaur… firefly.social/post/ff-54bc1b9f2fa1…

120

Vitalik Buterin @vitalik.ca · 20/07/2026

Then, we had a series of three eras of advancement, in which machines rapidly gained areas of capabilities that far exceeded our own: * In the industrial revolution, mass manufacturing, or generally the ability to perform… firefly.social/post/ff-7a6275281d35…

110

Vitalik Buterin @vitalik.ca · 20/07/2026

Thread: how I think about the growth of AI capabilities and what it means for us as they keep getting stronger. I think the right model for comparing humans and AI is not any single dimension like intelligence or… firefly.social/post/ff-47c3ae1478d5…

2120

Vitalik Buterin @vitalik.ca · 19/07/2026

Vibe-coded a toy demo version of the "anon billboard with moderation" idea from vitalik.eth.limo/general/2022/06/15… on aztec github.com/vbuterin/aztec_experiments It's early days but you can already do very interesting nontrivial stuff!

2112

Vitalik Buterin @vitalik.ca · 04/07/2026

Two weeks ago, Ethereum researchers met in Berlin to continue charting the protocol's long-term trajectory, following along discussions with client teams in Svalbard in April. The updated strawmap is at strawmap.org, and… firefly.social/post/ff-7e011098340f…

0194

Vitalik Buterin @vitalik.ca · 29/06/2026

A ten-thousand word monster post trying to cover the entire tech tree behind the main lineage of obfuscation (iO) protocols: vitalik.eth.limo/general/2026/06/29… Special thanks to all who helped!

1324

Vitalik Buterin @vitalik.ca · 11/05/2026

This theorem (left) means, the only way you can make proofs for two different things in the same position in the same Merkle tree, is by breaking the underlying hash function. As a reviewer, you don't have to verify how… firefly.social/post/ff-68d0355d8e98…

1101

Vitalik Buterin @vitalik.ca · 22/03/2026

x.com/Math_files/status/20352645856… I'm impressed that Qwen 35B was able to figure this one out. (took 15k tokens, a little under 3 min on my laptop) Anyway, the way that you can use trigonometric… firefly.social/post/ff-06158f292d0a…

9355

Vitalik Buterin @vitalik.ca · 12/03/2026

"Friendship ended with drawio, vibecoded SVG is my new friend"

2151

Vitalik Buterin @vitalik.ca · 21/09/2025

I agree! And I think low-risk defi is probably the single large-scale economic opportunity available to blockchains that doesn't create incentives to start bending ethically (whereas eg. political memecoins, high-risk ponzis, even HFT certainly do)

1100

Vitalik Buterin @vitalik.ca · 24/01/2025

On the risks of politician coins:

The risk of politician coins comes from the fact that they are such a perfect bribery vehicle. If a politician issues a coin, you do not even need to send *them* any coins to give them money. Instead, you just buy and hold the coin, and this increases the value of their holdings passively.

Furthermore, there is deniability: holding the coin is, in terms of financial effect, a linear combination of donating to the issuer and gambling. Hence you can have the intention of doing the former but when challenged claim that you are doing the latter.

You can even hold the coin privately, and show that you are holding it to whoever you need to show; you do not need any zero knowledge proofs, you just send a test transaction.

This is all risky to democracy, for reasons very similar to what I wrote in https://vitalik.eth.limo/general/2021/08/16/voting3.html , https://vitalik.eth.limo/general/2020/09/11/coordination.html and elsewhere. TLDR: the economic arguments for why markets are so great for "regular" goods and services do not extend to "markets for political influence".

I recommend politicians do not go down this path.

27754184

Vitalik Buterin @vitalik.ca · 24/01/2025

Reposting a large post from the other app.

There is perhaps an analogy with weed here.

Ten years ago, to many weed represented freedom, and rebellion against sclerotic old order that denied self-sovereignty over our bodies. Then, weed became legalized, and "official".

On that day, I remember my personal interest in weed dropping by > 2x overnight.

And since then, public discourse is moving toward a balanced view of weed: it is much less harmful than alcohol and smoking, and has valuable medical uses, but it is also far from harmless, especially in its modern corporate hyper-optimized versions.

Crypto finance is similar. In the old order,
@GaryGensler
created a regime where tokens that give investors clear indication of what their rights are and where their revenue comes from is a "security", but obfuscated "governance tokens" are potentially in the clear. This was a perverse distortion of incentives, and it is a key reason why
@GaryGensler
must never be christened as a hero, even among crypto skeptics. Everything that happened in crypto was in part a response, part compliance part rebellion, to these pressures.

Over the last year, we have been entering a new order. Now, the most powerful people in the world are cheering on the idea of anyone creating tokens for anything, at any scale.

And so now is the time to talk about the difference between sugar-high short-term fun that is unwise to recommend to newbies, and long-term fulfillment and wealth-building. It is not about "fun is bad", it is about the equivalent of modern hyperaddictive cellphone games, versus chess or World of Warcraft.

Now is the time to talk about the fact that large-scale political coins cross a further line: they are not just sources of fun, whose harm is at most contained to mistakes made by voluntary participants, they are vehicles for unlimited political bribery, including from foreign nation states.

This is in a sense the essence of d/acc: the goal is not to indiscriminately accelerate everything, including superintelligent AI and fentanyl, the goal is to selectively accelerate defense so that it can keep up with offense. Sometimes, that defense is a matter of cultivating our own sense of virtue. Sometimes, that defense is a matter of creating better alternatives.

In defi, we are already starting to see a return toward honest token-based fundraising that gives users clear understanding of what they are buying, bringing us the best of pre-2020-era ICOs while addressing their downsides. The
@infinex_app
patron sale, and the
@Truemarketsorg
NFT sale, are two examples of this, and I personally know of other examples that are soon to come.

There is a bright future of capital allocation mechanisms that can be built. Potentially, we can come up with ways to ensure alignment with community wishes as well as safeguarding important values like privacy, security, open standards and open source, that could be part of the incentive structure itself. Acceleration is coming either way; it is our task to choose the brightest possible vector.

510616

Vitalik Buterin @vitalik.ca · 14/12/2024

Introducing deep funding. Two key ideas: 1. Value as a graph: instead of asking "how much did X contribute to humanity?", ask "how much of the credit for Y belongs to X?" 2. Distilled human judgement: an open market of AIs fills in all weights, human jury randomly spot-checks them deepfunding.org

810214

Vitalik Buterin @vitalik.ca · 29/11/2024