As a friend said to me this morning: "Time to upgrade. Everything." 🎯
BREAKING: Google says “the era of autonomous attacks has arrived.”
Hackers are now using AI to scan software updates, figure out which security flaw each patch fixes and build an attack before companies can install the fix, all without a human involved, according to a new report from the Google Threat Intelligence Group.
“Defending against machine-speed threats at human speed isn’t feasible,” Google said.
3
"The whole point of these things is that they’re clumsy, inefficient, and carry the warmth of your own life."
Love that ❤️
刷完马斯克最新那场访谈,我持续了大半年的 AI 焦虑突然被治愈了,他直接甩出一个很反直觉的判断,
说未来人类的工作,大概率会彻底变成园艺。你想想看,今天你根本不需要在后院自己种番茄和土豆,
楼下超市买到的永远比你自己种的更快、更便宜、品相更稳定,但依然有大把人周末在阳台折腾花草,
甚至花一整个下午为朋友做一顿自己种出来的晚餐。
这件事在工业效率上毫无意义,可人类做这件事,本来也跟效率无关。
很多兄弟一聊到 AI 替代人类,第一反应永远是失业、断供和阶层滑落,但我仔细琢磨了一下,
AI 真正击碎的其实未必是饭碗,而是人类过去几千年赖以生存的价值坐标,也就是有用性。
过去几千年,一个普通人的尊严几乎全靠有用性撑着,你能搬砖、能算账、能写代码、能做报表,
社会按月给你发薪水,你靠这个确认自己被世界需要。
但转念一想,当 AI 在任何一项具体任务上都比人类做得更快、成本更低时,有用性瞬间就通货膨胀了,
机器负责解决生存的效率,人类反而被逼着退回生命本身的体验。
就像那些退休后过得极度充实的老人,每天早起练字、下棋、养花、钓鱼,没有一件事是在创造 GDP,可每个人都活得自得其乐。
说实话,未来所有能被标准化的实用技能,大概率都会被算法和机械臂彻底端走,
而那些效率最低、最耗时间、最不划算的事情,反而会变成精神上真正的奢侈品。
手写一封字迹歪扭的信,亲手烤一块火候不匀的面包,花整个周末陪孩子搭一个不讲力学的树屋,
这些事情的全部意义,就在于它们是笨拙的、低效的,并且留着你自己活过的温度。
如果未来靠打工换来的社会身份全部贬值,我先说我自己,写代码和做排版我秒甩给 AI,但给朋友做一顿慢吞吞的手工饭,我肯定是都要自己做的😄
56
People hand rolling code now are falling way behind (in delivering or in adapting to the times)
From the Anthropic Road Show video:
'This has also made us really good at bringing up new chips. As an example, this year, we started working with AMD GPUs for the first time. In the past, bringing on a new chip platform took dozens of engineers months of work. This time we had Mythos try it out. It got Claude running on AMD in just 3 days with basically no human intervention. Claude ported or wrote almost 100% of the kernels.'
1
33
"Welcome to the Agent Economy" 🎯📈
Agents now make up the majority of web traffic, write more than half of new articles published online, and generate most of the code pushed to production.
This is what the early stage of a platform shift looks like: adoption is measurable, but the business models it enables mostly aren’t visible yet.
Welcome to the agent economy.
a16z.news/p/welcome-to-the-b…
28
And it's barely started. Very few households pay for it yet, and a smaller percentage of businesses than you would guess. Huge room for growth in the decades ahead
Hard for me to look at these charts from the FT, especially the first one, and not see some evidence of “ROI on AI.”
Seems like a productivity surge post-Covid era overhiring that began to reverse before ChatGPT came out and then resumed its upward trend after ChatGPT.
Especially given the @a16z data on revenue per employee at their portfolio companies and the @EPBResearch data showing that the 3 industries that have most heavily adopted AI have gone from 2.2% productivity growth to 5% productivity growth.
And makes sense that it is showing up in the US first - America is 72% of Anthropic revenue vs. 50%ish at most application software companies.
Great stuff from @jburnmurdoch even if I disagree with his conclusion.
26
The primary atomic clock at NIST is so accurate that it will not gain or lose a single second in over 300 million years. This translates to a daily precision of within one quadrillionth of a second (0.000000000000001 seconds). ⏱️🎯
25
Interesting concept that seems useful at a glance. Curious to try this out for improved planning
Google just open-sourced the plugin thousands of its engineers use every day to stop AI from shipping sloppy code.
It's called the Elephant-Goldfish Model.
No code until the plan survives. No AI grading its own homework. No context collapse halfway through a feature.
The idea is simple and kind of brilliant.
The Elephant is one long-running AI session that remembers everything: the architecture, the intent, every small decision. It writes zero code. It interviews you, argues with you, and keeps the record.
The Goldfish are fresh AI sessions with no memory at all. They never saw your design conversation, so they don't inherit its blind spots.
Here's how it works:
→ egm-brainstorm: you and the Elephant debate options under a strict no-code rule
→ egm-draft: you produce a Markdown design doc with the problem, the plan, the alternatives, and the exact files to change
→ egm-review: one Goldfish has to explain the design back. If it gets it wrong, the doc is ambiguous and goes back for redrafting
→ Then several Goldfish attack it as hostile reviewers hunting edge cases, hidden dependencies and logic gaps
→ A final Goldfish makes the go/no-go call: are file paths, data structures and function signatures all present and unambiguous?
Here's why this matters:
Google's team says AI made coding faster but also sped up slop, bugs and session context collapse. Speed without guardrails just scales your mistakes.
This flips the order. The design gets attacked before a single line exists.
Here's the wildest part:
This isn't a research demo. Google says thousands of Googlers run it daily to ship production code.
The repo ships the Markdown rules, the reviewer personas and the review workflows. Free, Apache 2.0.
The honest trade-off: it only runs inside Antigravity for now, with other environments coming. And security isn't on the default reviewer list. You'd add your own security Goldfish.
github.com/google/elephant-g…
33
Thought provoking conversation with Karpathy.
Agents, when done right for tasks, don't need the biggest, most capable general model. They need evals around their functionality. Smaller, more purpose-built models can perform perfectly well with (most) agentic workflows
This is f*cking insane. Andrej Karpathy (ex-OpenAI, ex-Tesla AI lead):
"Agents aren't magic. They're distillation at scale
99.99% of your model's capacity is wasted on data it never needed"
In a 16-minute conversation, he lays out the stack that actually works: a small model, the right tools and a closed loop
Big Model → Small Model → Right Tools → Closed Loop → Self-Improving Agent
Most people keep paying for a bigger model. The loop is what makes a small one dangerous
Worth more than any $500 AI course this year
Bookmark and watch it today
Then read the "Harness, Loops, Graphs" guide below - the closed loop is the harness ↓
28
institute.deepmind.com/essay…
"Alongside individual leaderboards, we need benchmarks for organizations of agents: measuring how well multi-agent systems divide problems without duplicating work, share context cleanly across handoffs, & catch mistakes over long horizons."
24
Fascinating and rings true. Love the proof of work idea suggested
This paper answers why confident idiots run the world.
Cornell University published a psychological paper that explains every terrible boss, every loud influencer, and every broken corporate hierarchy.
It’s called “Unskilled and Unaware of It.”
they tested participants across logic, grammar, and humor to measure the gap between actual performance and self-perception.
And the results are wild.
participants scoring in the bottom 12th percentile consistently believed they were operating in the 62nd percentile.
they were not lying. they were literally incapable of recognizing their own incompetence.
the paper revealed a brutal paradox:
the exact skills required to master a domain are the exact same skills required to accurately evaluate your own competence in that domain.
if you lack the skill, you inherently lack the cognitive capacity to realize you are failing.
this creates a dangerous systemic loop.
the incompetent are blessed with absolute, blinding certainty. their brain protects them from their own ignorance, allowing them to speak first, demand authority, and never second-guess their decisions.
meanwhile, the actual experts suffer the exact inverse.
top performers in the study consistently underestimated their rank. because a task was easy for them, they falsely assumed it was easy for everyone else.
they hesitate. they obsess over nuance. they assume they have nothing unique to add.
we are biologically wired to confuse confidence with competence.
the loudest person in the room gets promoted, gets funded, and gets the microphone, not because they know what they are doing, but because they are physically incapable of seeing what they don't.
the only defense mechanism is building systems that stop rewarding absolute certainty and start demanding proof of work.
38
Recently earned my Improving AI Stage 4 certification, which got me reflecting on intrinsic vs. extrinsic motivation.
It's the learning that compounds, through the curiosity, growth, and capabilities we carry forward.
holtonma.github.io/posts/the…
35
Considerations for where to draw the line with sub agents
claude.com/resources/article…
2
45
Times have changed... good to pay attention to this trend if you host anything
The window to patch software bugs is collapsing
Of the bugs hackers actually exploit, ~87% are now being attacked on or before the day the bug is public knowledge
That share was 23% in 2020
Charts of the Week: a16z.news/p/chart-of-the-wee…
1
43
Exciting how rapidly this is evolving
Karan Singhal of @OpenAI on the collapsing cost of intelligence, as seen on their health benchmark:
"GPT-6 Luna is our current cheapest frontier model on the API... It performed better than 5.6 Sol, which was literally our best model a month ago. And it performs better despite being 34x cheaper."
Most tiers of model intelligence became more than 100x cheaper within roughly a year
One tier went from ~$32 to ~$0.02 per million tokens in two years
1
33
Whether you use decision models or not for evals is an evolving question, but certainly should be building evals for your agents, workflows, prompts
Learn to build evals, folks.
Highly recommended reading on using decision models for evals.
From our internal evals, System One models like Jev are a great fit for LLM Judges.
If you want to go beyond Hamel's great write-up, I've also put together a few resources on applying decision models to LLM judges.
(bookmark it)
1) Jev-as-a-Judge for Agent Evaluations - academy.dair.ai/dashboard/re…
2) Jev-as-a-Judge: Evaluating Agents with a Decision Model - academy.dair.ai/labs/jev-as-…
Bonus paper I would recommend, with a few other ideas that combine System One and System Two models: Jev-as-a-Judge paper - arxiv.org/abs/2609.26550
1
38
If you are building Skills and want consistency in prompts (and skills), PromptFoo is great!
github.com/anthropics/course…
57
Bedard coming back last night to the #Blackhawks , and the way the PP operated has given me *hope* this year.
I need to be talked down like Redd did to Andy Dufrain in Shawshank.
"Lemme tell u my friend, hope, is a dangerous thing... "
1
3
90
We are seeing this more and more (I blogged about @cognition previously and their RL for coding, where they RL on top of KimiK open weight models). This pattern will gain steam it seems, as it delivers a lot of value without a (relatively) massive cost
We were able to fine tune an open weight model (Qwen3.5-9B) on proprietary internal data at @coinbase, and it out performed frontier models, at much lower cost and speed, for a specialized task (fraud prevention in this case).
I think we'll see more of this pattern inside companies. The frontier models are amazing generalists. But increasingly you'll see specialist models out perform on more narrow tasks, either working on teams, or by themselves.
We're making it easy to fine tune models internally for our employees. We even have a skill for it in our internal harness that makes it easy to spin these up and start training them:
"/fine-tune-and-serve-llm I want to fine-tune Qwen for payments risk review. help me get started"
I imagine we'll have some backend services that could be replaced entirely by this new architecture.
43
Token prices keep falling and GPU rental prices keep rising
Intelligence gets cheaper, usage goes up, and a model ends up in every piece of software
TypeSafe AI's Diogo Almeida on why Jev is called Jev, and what cheap intelligence finally changes about software:
"It's wild that AI is so cool and software has been unchanged in 10 years... The most we can do is add a little chatbot in the side sometimes that can take actions, but not all actions."
"TypeSafe is making AI for software. We want to make AI powerful, not just for humans in the loop, but to actually build real software."
"It's about the laziness to spend 10 hours to do the five-minute task instantly and to never have to do it again. It should be an ROI decision for people who automate stuff. I would just like it to be automatable."
"I think that people will just make new kinds of work, hence the Jev in Jevons. New kinds of work once that stuff is doable."
"As we have lots of intelligence at lots of different cost and speed trade-offs, the super systems-y types will be making trade-offs. Jev's gonna be 1000x too smart for them. They just want an approximate guess to optimistically route here and there."
Martin Casado: "The good news is we get to rebuild systems again. We have a new primitive. It's kind of a new way of thinking about doing software."
@CompleteSkeptic @typesafeai @bhorowitz @martin_casado
























