Vitalik: We Shouldn't Oppose AI, But Create Sanctuaries

marsbitPublished on 2026-05-17Last updated on 2026-05-17

Abstract

Summary: In a podcast, Ethereum co-founder Vitalik Buterin discusses the human response to advancing AI. He argues the core risk isn't AI's intelligence, but humanity's passive reliance on centralized systems and AI, which erodes privacy and personal agency. His solution is not to fight AI, but to build "sanctuary technologies"—tools like Ethereum that offer safe, optional spaces where individuals can retain sovereignty and freedom of choice without being forced into a single system. Buterin reflects on his shift from an "autopilot" mode in his youth to actively "piloting" his life and work. He stresses the importance of maintaining "manual mode" activities—like learning actively and performing tasks without AI assistance—to prevent mental atrophy and preserve independent thought. For builders, his advice is to prioritize technologies that protect human agency, avoid outsourcing all strategic thinking, and actively engage in serendipitous, real-world interactions. He repositions Ethereum/crypto not as a fix for legacy systems like the dollar, but as a parallel, voluntary option. In an era of increasing centralization, the freedom to opt-in or opt-out becomes invaluable. The key takeaway is that the AI era demands greater human proactivity; the scarcest resource will remain actively thinking, sovereign individuals.

Author: Saito

Just finished listening to this episode of Vitalik's podcast with a16z, and the amount of information was mind-blowing.

He founded Ethereum at 19 and is now in his early 30s, having transitioned from living on "autopilot" to "actively piloting."

The core topic of this episode is the question we are most anxious about today: AI is getting stronger, so what should humans do?

Vitalik's answer is not to "oppose AI," but to create sanctuary technologies. These are technologies that protect us without stripping away our privacy and agency (sovereignty).

Today, I'll break down the hardest counter-intuitive points, practical advice, and Ethereum's new positioning from this episode.

The Biggest Risk in the AI Era Isn't That AI Is Too Smart, But That Humans Are Too Passive

Vitalik bluntly stated that the world today is less secure and less peaceful than it was 10 to 15 years ago.

Many people are pursuing a kind of "safety": handing everything over to "Uncle in the Sky," meaning big companies, super AIs, centralized systems, letting them make decisions for us, manage risks, and provide protection.

But the cost of this safety is that we lose privacy and lose agency.

Vitalik calls this kind of safety disempowering safety, safety that increasingly weakens people.

This is also where he reinterprets the mission of crypto / Ethereum. The significance of Ethereum is not to "fix the dollar," not to repair the existing financial system, but to create a new option. You are free to choose whether to use it or not.

This is what a true sanctuary is: both safe and allowing you to retain sovereignty.

Sanctuary Technologies: Small Spaces Preserving Human Freedom

Sanctuary Technologies is a term coined by Vitalik himself, and "庇护所技术" (bì hù suǒ jì shù, sanctuary technology) is a fitting translation.

It's not about turning the whole world into a safe house, nor is it about ruling everyone with a larger system. What it truly aims to do is: give you a safe, small space where you can think, coordinate, and create freely without being completely controlled by external forces.

It has several core characteristics: it is not totalizing, it does not attempt to rule the entire world; it preserves privacy and agency; everyone can freely enter and exit, it is not mandatory.

Ethereum is a typical sanctuary tech. It doesn't try to fix the existing financial system; it gives you a parallel option. Use it if you want, don't if you don't.

This will become increasingly important in the AI era. Because as big companies and super AIs become stronger, what humans truly need is not another system that "arranges everything for you," but a space that preserves your right to choose.

From Autopilot to Active Pilot: Vitalik's Personal Growth

Vitalik reflected that when he founded Ethereum at 19, he was largely in an autopilot state.

Many decisions were made by going with the flow: dropping out, writing the whitepaper, being denied a visa by Ripple, which instead became a turning point in his life. Back then, he was more like being pushed along by the world.

But now he is increasingly realizing: the world changes too fast, no one is coming to save you, you must be the pilot yourself.

He gave a few very relatable examples. Ten years ago, people could go days without contacting friends; now, not replying to a message for a day causes anxiety. Ten years ago, you could actually get "lost" while walking; now, with phone navigation, cities have become a series of "teleportation points."

These changes remind us: the world "dies and is reborn" every 5 to 10 years. If you keep living by the old script, you'll fall behind quickly.

So what's truly important in the AI era is not passively waiting for technology to take you somewhere, but actively deciding how you will use technology.

The Stronger AI Gets, The More Humans Must Keep "Manual Mode"

Vitalik specifically emphasized: active learning is 10 times more effective than passive learning, even if you spend the same amount of time.

From a young age, he would force himself to do many things manually, like not using a calculator in chemistry class or not using navigation while walking. The goal wasn't to be anti-technology, but to keep his brain engaged.

The stronger AI gets, the more we should deliberately retain some "manual mode."

Sometimes deliberately not using AI to write code, sometimes deliberately walking without navigation, sometimes deliberately not letting a chatbot think through problems for you.

This isn't nostalgia, nor is it rejecting efficiency; it's to prevent brain atrophy and maintain one's own agency.

AI can help us do many things, but if all thinking, judgment, and exploration are outsourced, people will slowly become passengers in the system. Vitalik's reminder is: you can use AI, but don't let yourself become completely dependent on it.

Practical Advice for Builders

In this episode, the inspiration Vitalik gives to ordinary builders is very direct.

First, force yourself to do things manually. Even if AI can help you, occasionally do things yourself to ensure your brain doesn't get rusty.

Second, active learning. Don't just let AI give you answers; deduce, verify, and do things yourself.

Third, build sanctuary technologies. Whether you're building open-source tools, decentralized protocols, or a personal knowledge base, prioritize one thing: does it help people retain sovereignty?

Fourth, don't outsource all your brainpower. AI can help you with execution, but strategy, direction, and values must be controlled by you.

Fifth, maintain serendipity. Participate more in offline events, talk more with real people, don't leave all discoveries to algorithm recommendations.

These points all essentially point to the same core: the AI era isn't about using tools less, but about using tools more actively.

Ethereum's New Positioning: Not Fixing the Old World, But Creating New Options

Vitalik is also clear about crypto's positioning.

Crypto can't solve all of the dollar's problems, and it doesn't need to pretend it can solve all problems. But it can create something new without those shortcomings.

Everyone can freely choose to use it or not.

This is crypto's greatest strength: it doesn't force you; it gives you choice.

In an era where AI power is increasingly concentrated, this will become increasingly precious. Because as more and more systems try to make decisions for you, filter information for you, and judge risks for you, a parallel option that is non-coercive and allows free entry and exit becomes very important in itself.

The value of Ethereum / crypto is not "beating the old world," but giving you a new world you can freely choose.

The Most Counter-Intuitive Lines from This Episode

The biggest risk in the AI era is not AI replacing humans, but humans willingly turning themselves into passengers.

A sanctuary isn't about making the whole world safe, but giving you a safe, small space where you can still retain freedom.

Active learning is 10 times more effective than passive learning, even with the same time spent.

The world dies and is reborn every 5 to 10 years; we must be our own pilot.

Inspiration for Ordinary People

The stronger AI gets, the more proactive humans must be.

Don't outsource all your thinking to models. Do more manual things to keep your brain engaged. Participate in building tools that preserve human sovereignty, whether open-source, decentralized, or personal knowledge management systems.

Remember: technology ultimately serves humans, it does not replace humans.

Vitalik concluded by saying that we humans are the brightest stars. AI can be strong, but what truly drives the world forward are still active, agentic people.

Summary in One Sentence

With his 10 years of personal experience, Vitalik tells us: The AI era is not an era to lie back, but an era that requires humans to actively pilot more than ever.

Don't outsource your brain to models. Do more manual things, build sanctuary technologies, preserve your privacy and agency.

My biggest takeaway after listening to this episode is: Before, we feared AI taking our jobs; now it seems AI is actually upgrading people from "executors" to "designers."

What's truly scarce has never been computing power, but people willing to think actively and preserve their sovereignty.

Related Questions

QAccording to Vitalik, what is the main approach we should take towards AI, and what is the core concept behind it?

AThe main approach is not to fight against AI, but to create 'Sanctuary Technologies'. The core concept is to build safe spaces that protect humans without stripping away their privacy and agency (sovereignty).

QWhat does Vitalik mean by 'disempowering safety', and why is it a risk in the AI era?

A'Disempowering safety' refers to the kind of security gained by handing over decision-making, risk management, and protection to big companies, super AIs, or centralized systems. The risk is that while it offers safety, the cost is the loss of human privacy and agency, making people increasingly passive.

QBased on the article, what is the new purpose or positioning of Ethereum/crypto as explained by Vitalik?

AVitalik repositions Ethereum/crypto not as a tool to 'fix' or replace the old world (like the current financial system), but as a creator of a new, parallel option. Its power lies in offering people a voluntary choice—to use it or not—thereby providing a sanctuary where one can retain sovereignty.

QWhat are the practical suggestions Vitalik offers for builders and individuals in the AI era?

A1. Force yourself to do things manually sometimes to keep your brain active. 2. Engage in active learning (self-derivation, verification) rather than just accepting AI answers. 3. Build tools and technologies that help preserve human sovereignty (sanctuary technologies). 4. Don't outsource all strategic thinking and values to AI; maintain personal control over direction. 5. Keep serendipity alive through real-world interactions and not relying solely on algorithmic recommendations.

QWhat is the 'most counter-intuitive' point Vitalik makes about learning and human agency in relation to AI?

AHe states that 'active learning is 10 times more effective than passive learning, even if the time spent is the same.' This underscores the importance of maintaining manual, self-directed thinking and agency to prevent mental atrophy, especially as AI grows more powerful.

Related Reads

The Revelation from the Raydium Theft Incident: New DeFi Vulnerabilities Lurking in Forgotten Old Contracts

**Raydium Exploit Reveals DeFi's Hidden Risk: Forgotten "Zombie" Contracts** A recent attack on Raydium's deprecated V3 AMM pools resulted in a loss of approximately $1.34 million. The hacker exploited pools that were no longer supported by Raydium's current UI or SDK but remained fully functional and accessible on-chain. This incident highlights a critical, often overlooked category of risk in DeFi: inactive or legacy smart contracts that projects fail to properly decommission. Since March 2025, there have been at least 8 publicly reported attacks targeting such abandoned contracts, with total losses around $10.8 million. Including older pools and deprecated features, the count rises to 10 incidents with roughly $22.5 million in losses. These "zombie contracts" represent a lifecycle management failure rather than a code vulnerability, yet they are typically misclassified under general "code bug" categories in security reports, masking the true scale of the problem. The root cause is that projects often merely document a contract as "deprecated" without taking essential technical steps to secure it: withdrawing remaining assets, disabling external call functions, and implementing ongoing monitoring. These forgotten, under-monitored components become prime targets for attackers. To address this, the industry needs to recognize "zombie contracts" as a distinct risk category and establish standardized decommissioning protocols. Essential steps should include: 1) a formal retirement announcement, 2) removal of all front-end integrations, 3) withdrawal of locked assets, 4) disabling key contract functions, 5) ongoing security monitoring, 6) clear user communication, and 7) a post-mortem analysis. The value of a DeFi project lies not only in its current TVL but also in the security of its historical codebase, which has now become a new attack surface.

Foresight News34m ago

The Revelation from the Raydium Theft Incident: New DeFi Vulnerabilities Lurking in Forgotten Old Contracts

Foresight News34m ago

Robots Begin to 'Consume Data': The Hidden Production Chain from Indian Data Factories to Billion-Dollar Humanoid Robots

Robots have started to 'consume data,' driving the formation of a new industrial supply chain focused on producing training data for embodied AI. Unlike large language models, which are trained on vast internet text corpora, embodied AI models face a 'data desert' in the physical world. This has created a massive demand for first-person perspective video data (Ego Data), captured by workers wearing cameras in places like Indian garment factories. Companies like Neocambrian AI are establishing 'data factories' where workers perform standardized tasks (e.g., sorting clothes, kitchen organization) to generate thousands of hours of video. Research, such as NVIDIA's EgoScale, demonstrates that scaling this human demonstration data predictably improves robot performance, particularly for dexterous manipulation. This has validated a training path combining large-scale human data for pre-training with smaller amounts of robot-specific data for fine-tuning. The value of different data types varies significantly, forming a 'data pyramid.' The base consists of low-cost, large-scale internet and Ego Data. Higher layers include more expensive motion-capture data (e.g., from data gloves), simulation/synthetic data, and the most costly and scarce layer: real robot teleoperation data. This demand has spawned a layered ecosystem of data suppliers: low-cost data factories, motion capture and alignment specialists, robot-native teleoperation service providers, simulation data companies, and platforms aiming for data standardization. Robot companies themselves are adopting a 'layered procurement' strategy: outsourcing generic Ego Data while building in-house capabilities for robot-specific adaptation data and the critical deployment/failure data generated in real-world applications. The industry is shifting focus from hardware and basic mobility to the data pipelines required for general-purpose capability. While parallels exist to data labeling companies like Scale AI in the LLM boom, the physical complexity of robot data—involving action success ambiguity and sim-to-real gaps—requires more integrated solutions for data collection, annotation, and a continuous feedback loop. The race is on to build the data engines that will teach robots to operate reliably in the unstructured real world.

marsbit3h ago

Robots Begin to 'Consume Data': The Hidden Production Chain from Indian Data Factories to Billion-Dollar Humanoid Robots

marsbit3h ago

Spicy Commentary | Michael Saylor's 'Player Talk'; 60-Year-Old Aunt Liquidated After 'Scamming a Young Man'

**"Spicy Commentary": Three Tales of Crypto's Wild Week** This week's "Spicy Commentary" column highlights three dramatic stories from the cryptocurrency world. First, **MicroStrategy's Michael Saylor** addressed the controversy over his company potentially selling Bitcoin. At the BTC Prague event, he clarified, "I never said the company can't sell Bitcoin. I told *you* never to sell *your* Bitcoin." This "do as I say, not as I do" stance was criticized by netizens as peak linguistic gymnastics, noting a history of him previously stating the company would "never" sell. Second, a **bizarre fraud case** emerged from Beijing. A 60-year-old woman, obsessed with getting rich from crypto but unwilling to risk her own savings, posed online as the 20-something "god-daughter" of a high-ranking official. She catfished a young man, convincing him to give her over 200,000 yuan for fabricated emergencies. She then invested all the stolen money into cryptocurrency with 10x leverage, only to lose everything in a market crash. The woman was sentenced to four years in prison for fraud. Finally, a **sobering trader's tale** surfaced on Reddit. A user posted "Tale of a crypto trader," confessing their net worth had plummeted from a peak of $45 million to roughly $17,200, primarily due to holding meme coins too long. The post, described as a crypto "book of confessions," sparked reactions ranging from sympathy to critique about greed, poor risk management, and the perils of treating meme coins as long-term investments instead of taking profits. The column concludes that this week featured masterful rhetoric, elaborate scams, and extreme financial volatility, stitching together another chapter in crypto's unpredictable theater.

Foresight News3h ago

Spicy Commentary | Michael Saylor's 'Player Talk'; 60-Year-Old Aunt Liquidated After 'Scamming a Young Man'

Foresight News3h ago

Tremble Humans, AI Continues Its Accelerated Sprint

Trembling, Humans: AI Continues Its Accelerated Sprint Yes, AI is still rapidly accelerating. While deep learning seemed to stall quickly in its early years, large models after years of development show no sign of hitting their ceiling. At the Zhiyuan Conference 2026, the focus is on enabling AI to move from the digital world into the physical world. Scaling Law remains effective, continuing to drive advancements in both large language models and multimodal models. The industry is now entering a phase of pursuing World Models, though unresolved technical paths and data issues mean this exploration may take 3-5 more years. Concurrently, breakthroughs in Agents are accelerating AI's real-world application in fields like healthcare and meetings. Making Agents truly useful requires key hardware-software co-design, evident from the strong presence of chip vendors at the conference. We stand at a new historical threshold where AI is becoming a foundational force reshaping the world. The first day of the conference highlighted AI's evolution from "knowing how to chat" to "knowing how to work." Scaling Law persists, World Models are the next key battleground, and Agents are transitioning from usable to好用 (user-friendly). Scaling Law is not ending but diversifying. New models like Anthropic's Fable 5 demonstrate scaling through parameter size, synthetic data, and reinforcement learning. Advancements in AI Coding and Agent deployment are enabling a trend of AI self-evolution, potentially allowing AI to take over digital world iterations. World Models represent the next frontier for large models extending into the physical realm, but no current model is truly impressive at solving real-world problems. Technical consensus is lacking, with debates on data sources (video, simulation, real-world). Different approaches are emerging: language-centric, pixel-centric, 3D-structure-centric, and visual-representation-centric models. Zhiyuan Institute is exploring a fifth path: unified latent space modeling fusing language and visual representations, and introduced its own under-development World Model, Physis-v0.1. On the product side, Agents are key to bringing AI into daily life. Since 2025, the "Year of the Agent," products have become more proactive and capable of complex tasks. Zhiyuan showcased four vertical Agents for cardiac diagnosis, autonomous research, meeting summarization, and protein risk discovery. However, technical challenges remain, particularly in context engineering like memory and orchestration. "Harness" – the engineering framework around an Agent – is crucial for maximizing its capabilities by clarifying intent, designing workflows, and incorporating validation and feedback. In summary, AI's breakneck pace continues on multiple fronts: foundational model scaling, the ambitious pursuit of World Models for physical understanding, and the ongoing refinement of practical Agents. The journey from capable to truly reliable and useful AI systems is well underway.

marsbit3h ago

Tremble Humans, AI Continues Its Accelerated Sprint

marsbit3h ago

Trading

Spot
Futures

Hot Articles

Discussions

Welcome to the HTX Community. Here, you can stay informed about the latest platform developments and gain access to professional market insights. Users' opinions on the price of AI (AI) are presented below.

活动图片