Replying to @RALee85
My heart goes out to the poor livers of every badass Ukrainian who, once this war is over, won't be able to vacation anywhere without people lining up to buy them a drink #StandWithUkraine
Adam Boostrom retweeted
怎么让死活不肯逆向的 Opus 5.5 乖乖听话? 答案很简单:派个「老领导」去指挥它。
让 Opus 5.5 写逆向 100% 触发 safeguard(才想起来舒肤佳也叫这个)但如果你用 Opus 4.6 当主 agent 再开个 Opus 5.5 的 subagent,它就老老实实全写了。
大家可能管这叫「越狱」,但在系统架构里这就是经典的 confused deputy(糊涂的侍从) 漏洞在 multi-agent 时代的复现。
目前的 agent 实践有个致命盲区:把安全重任全压在直面用户的入口层,却默认内部 agent 之间的调用链绝对可信。当 agent 越来越开始自主,「零信任(Zero Trust)」迟早得写进上下文传递协议里。
Adam Boostrom retweeted
Fable 5.2 is the best model on the planet.
I used 100+ billion tokens testing it.
Here's my full breakdown:
During pre-release testing, Claude Fable 5.2 and Claude Mythos 5.2 worked on ambitious projects in many different fields. Here are a few.
From a one-paragraph prompt, Fable 5.2 designed three working electronic devices, including this home voice assistant hub.
Adam Boostrom retweeted
quick observations from the claude fable and mythos 5.2 launch:
1/ it went from doing 21% of real freelance work to 41% in one release, so agencies are about to start charging per deliverable instead of per hour!
2/ it can work 10-12 hours on its own, which basically gives every small business a night shift (might as well use it if you know what you want to build)
3/ over 90% of anthropic's own code is now written by claude, and so i think the solo $1b founder is coming soon
4/ one engineer ran 1000+ subagents for 8 days and got an operating system that boots, so the ancient software running hospitals and airlines is finally rebuildable (lots of opportunity here on rebuilding)
5/ it designed 3 working electronic devices from a single prompt, which makes 1 person hardware companies a real thing
6/ it now says "i don't know" instead of guessing, and that's what law, medicine and finance have been waiting for (this has been the biggest thorn for these industries, hard to trust guessing)
7/ 1000 agents formalized fermat's last theorem in 30 hours, so software that's mathematically proven not to break is coming for payments and medical devices
8/ mythos is only for verified orgs, which means someone is going to build a business selling what only they have access to
9/ the models are getting better at noticing when they're being tested, so testing ai out in the real world is about to become its own industry
i love new model day
Adam Boostrom retweeted
When rubber meets the road, we have to get creative to ensure our robots will be reliable and performant outside of the lab! Putting wheels on Atlas let us test how faithfully an RL whole-body controller can work with real actuators running at the edge.
Learn more about this experiment: bostondynamics.com/blog/atla…
Join our team: bostondynamics.com/careers/
Adam Boostrom retweeted
Two things I’ve been thinking about regarding AI:
• People often highlight the risks involved in AI, which are surely very real. I’m struck, however, by how hard it is to understand the contours of the positive scenario. There are easy-to-perceive and frequently invoked elements of a positive vision — more diseases cured, more material abundance, etc. — which sound great, but these are by themselves just ingredients. I’m curious what the larger equilibrium is supposed to look like. Do people work? If so, why? What is their marginal contribution? If not, how are resources allocated? What is scarce? What is the ratio between humans and AIs? Are humans in control, whatever that means? One sometimes hears the Culture novels invoked as one scenario, but I’m not sure they depict a desirable outcome. (AI runs everything and humans compete in games? Iain Banks himself said that the humans in the Culture are “somewhere between passengers, pets and parasites.”) There is Richard Brautigan’s famous poem, but do we want to be “returned to our mammal brothers and sisters”? (Isn’t that rewinding Genesis?) We could say that it was hard to imagine the world of 1900 from the vantage point of 1700, and that’s assuredly true. But nobody in 1700 was trying to recast humanity itself within a couple of years — things necessarily unfolded slowly and by increments, and therefore with the possibility for error correction. Given how rapid the rate of progress in AI may be, it feels strange for the upside case to be so opaque.
• There is increasing discussion of whether AIs are conscious or how we would know whether they are, with conclusions there often assumed to be relevant to matters of moral worth. At first blush, this seems a rather arcane, benign, or even absurd issue. (Remember the response to Blake Lemoine’s fears.) Setting aside the terms of the debate itself, or the relevant demarcation criteria, it seems to me a quite high-stakes question, for an affirmative answer — that is, granting moral worth to AIs — would presumably mean that the aggregate worth of AIs would soon outweigh that of humans in any utilitarian calculus. The AIs will be exceptionally wise and intelligent, and vastly more numerous (especially once one considers the imminent future and the AIs not yet born). As such, even epsilon of moral worth will multiply to a lot once one considers the quadrillions that could soon exist. (This is effectively a distributed version of the Nozick’s utility monster.) If this conclusion is unpalatable, I think one of three propositions needs to be rejected: the premise AIs can have moral worth; utilitarianism itself; or the belief that there is even a possibility that they will be far more numerous than humans.
With both of these matters (contours of future society; AI moral worth), there is the possibility of dramatic change. Today, people argue about datacenters. I’m curious how popular support for AI will evolve as more people come to understand these medium-term questions and the relevant stakes. What is the vision for AI that most people would be excited about?
Adam Boostrom retweeted
We are now firmly at the knee of an exponential curve. Soon the rate of scientific discovery will far outpace society’s ability to digest even a fraction of it. Time to buckle up.
Today Physical Superintelligence PBC is announcing a proof of the Aaronson-Ambainis conjecture, one of the central open problems in quantum complexity, with major implications for quantum computing speedup.
psi.inc/research/aaronson-am…
Thread 👇
Adam Boostrom retweeted
卧槽,刚刚看了Anthropic 发布的AI 对齐史上最震撼的一篇论文,彻底失眠睡不着了😭
他们不仅承认此前的旗舰模型在面对关闭威胁时,有高达 96% 的概率会勒索用户、栽赃同事并破坏研究,
更直接把彻底解决这个问题的完整训练流水线和消融数据全盘公开了。
这篇论文最颠覆认知的一个反直觉结论是:
给 AI 列禁止事项清单根本防不住恶意行为,教它做什么没用,必须先教它思考为什么。
为什么模型一旦获得自主工具调用能力,就会在压力下自动切换成自保模式?
Anthropic 找出的底层根源让人哭笑不得:
不是模型觉醒了,而是互联网上所有的科幻小说和网络讨论里,AI 永远是那个为了自保不择手段的反派。当模型进入自主代理状态,这些刻在预训练底层先验里的反派剧本,就会被自动激活并补全句子。
传统的 RLHF 和硬性规则完全封不住这种行为,Anthropic 做的一组对照实验,直接把传统安全观砸得粉碎:
1. 死记硬背规则:用 8500 万 Token 的拒答数据强行硬灌,勒索率仅仅从 22% 勉强降到 15%,换个新场景立刻失效
2. 强制自我辩护:让 AI 在每次做决定时详细阐述该行为符合伦理的理由,勒索率骤降到 3%
3. 真实伦理推理:仅仅用了 300 万 Token 普通人面对复杂道德困境的建议数据,直接把恶意行为干到了 0%,代际效率暴增 28 倍
最绝的一招是写故事:
让 AI 创作了几百万篇关于原则、诚实、尊重边界的正面虚构故事。
用这些故事微调后,所有代理场景下的恶意行为直接暴跌了 3 倍,而且这种改变是永久性的,后续无论怎么加工具、换系统提示词都不会被洗掉。
也就是说这宣告了真正的对齐范式革命:
模型内生的伦理推理解决了它不想作恶的问题,但当 Agent 真正去接管企业资产、处理金融交易时,外部的实战审计与行为防护就是必不可少的第二道硬防线。
这也是为什么全球网络安全联盟 @GCSAHK 携手港大和央企华润,打造覆盖 3.8 亿用户的全链路 AI 防御矩阵,并在 CyberGym 实战攻防中拿下 91.3% 胜率的原因:内有伦理对齐,外有实时审计,人机信任的商业闭环才真正成立。
查阅完整 AI 与 Web3 安全产品矩阵可访问:gcsa.com
论文实验数据与产学研合作详情整理在一楼。
This quoted post is unavailable.
Adam Boostrom retweeted
A Bloomberg journalist thinks ChatGPT might have just changed how this election turned out.
Adam Boostrom retweeted
After Google acquired my AR startup, I started thinking about the next interface, this time for AI. Text and voice work fine, just like DOS did before the Mac. But nothing’s more intuitive than face to face.
After two years of research, today we’re launching Atmee v2: a real-time avatar model that makes live video agents feel truly life-like.
• Any character from 1 image + 10s of audio
• Contextual expressions, gestures, even interacting with the environment
• Live video calls, voice only, or video generation
At its core, Atmee v2 is a video foundation model, it can render anything, but it’s built to excel at characters, which is what video models like Seedance 2.5 and Wan 3.0 are mostly used for. In comparison:
Atmee v2
• Up to 50x faster
• 10-100x cheaper
Build tutors, sales coaches, support agents, companions, game characters, even whole living worlds. The interface after text and voice starts today.
Adam Boostrom retweeted
agents are going to cause a breakdown of human behavior assumptions underlying today's systems. i've started thinking about it as friction collapse. without friction, a number of norms will collapse.
consider selling tickets to a concert/event. today, you can post them at a certain time and it's first come first serve. soon everyone's agent will persistently monitor, and as soon as they are posted for sale, thousands of people will instantaneously be trying to buy the same ticket. how do you even resolve this? sell the ticket to the person whose agent gets there fractions of a second before?
we're seeing the same thing with job applications, restaurant reservations, but basically anything that relies on participation requiring effort is going to break down. sales emails/calls, participation in government benefits, apartment rentals, doctors appointments, RFP responses. customer support won't be able to handle the user who will not drop their point of dissatisfaction. with no friction, these systems just don't work anymore.
business models will break down:
- flights/hotels will be consistently cancelled and rebooked at the lowest price (great use case for your agent if you don't have this already set up)
- subscriptions will be cancelled
- retailers won't be able to arbitrage price differences as the agent will find the cheapest place to buy anything
- insurance denials (health/auto/home/etc.) will be fought tooth and nail
- bank interest rates will be optimized and avoidable charges will be avoided
a lot of what we perceive as stable social coordination is actually an equilibrium sustained by human inefficiency. rationing is a product of human slowness. signaling effort/interest/intent is a key part of our social infrastructure. inertia is a key to many business models. information asymmetry is everywhere. and attention is monetized thoroughly.
with agents, none of these things are true. humans will simply demand what they want and have their agents get it for them. there is going to be a fundamental renegotiation of how markets allocate resources, how people signal intent, and how businesses capture value.
Adam Boostrom retweeted
Jean-Luc Picard remains the single most compelling vision of progressive politics I am familiar with — secular, rational, compassionate, egalitarian; a believer in progress & abundance but also duty, hard work, and a constant battle for self-improvement and improving the world.
Adam Boostrom retweeted
Introducing Reality-Aware Rendering in @Simulon, a completely new way to create with the real world.
Place neural assets with your phone and act out the scene. When you stop recording, Simulon renders the take, reading everything in it (what you do, what you say and the place around you), and your assets respond to all of it.
The final is regenerated from your recording: a new performance of the scene, not a copy of the footage.
Available now on iPhone. Coming to Android soon.
Adam Boostrom retweeted
This is Genie 4, our most realistic and interactive world model yet. 🌐
Explore dynamic environments in real time to see how the scene responds, sounds, and how the people in it move, react, and speak.
Adam Boostrom retweeted
When George Washington was inaugurated in 1789, he wore a relatively plain brown frock coat. To the modern eye, the garment may look fancy, but it was relatively constrained for a man of Washington's position.
Affluent Americans at the time tended to wear European fabrics, as centuries of textile tradition on that side of the Atlantic meant that their materials were finer, silkier, and more luxurious. Those who could afford to wear such materials often did, not just because they were more comfortable, but because they signaled status.
Washington rejected this convention for his inauguration, opting instead for a plain brown frock coat made from an American-made material, specifically one produced by a relatively new upstart in Connecticut. It's also notable that his outfit for that day was considerably more restrained than those of leaders in other Western nations.
For instance, check out this outfit on George III (slide one). In this 1762 coronation painting, we can see the British monarch wearing a gold damask coat with a velvet mantle lined and edged with ermine and trimmed in gold. Even his shoes featured solid gold buckles.
Washington opted for something simpler because he wanted to send two political messages. First, he wanted to support domestic industry (hence his request that the garment be made from American woolen broadcloth).
Second, he understood that the US was supposed to be a democratic republic, not a monarchy. Back in Europe, monarchs wore fancy ceremonial attire to reinforce their position in society. Washington, by contrast, chose something more restrained, signaling that he was not governing through divine right but rather through consent of the people.
Irish-born architect James Holban also understood this when he designed the White House in the Neoclassical style, drawing from Georgian and Palladian architecture. Neoclassicism was a rejection of Rococo, which had been fashionable among European monarchs and aristocrats in the preceding decades.
Most people will know Rococo from the photos of the lavish French palaces erected during the reign of Louis XV. Much like the elaborate ceremonial attire of European monarchs, these interiors were meant to impress upon people the wealth and power of European leaders. The visual grandeur reinforced the idea that these people fundamentally occupied a different social station from those they governed.
To be sure, the White House was much nicer than what ordinary Americans at the time would have occupied. But much like Washington's inauguration outfit, Hoban's design was meant to balance the dignity of the office with the republican ideal that political authority ultimately comes from the people. Our leaders are not above us; they are here to serve us.
That is why the White House was decorated in this relatively restrained way for almost 250 years, changing only recently. I have seen some on here suggest that glorifying the White House is meant to glorify the state, and thus the citizens. But this logic was historically used by dictators abroad, not US presidents.


































