今日科技推特头条 - 2026年5月26日

2026年5月26日科技每日简报

Today's top tech conversations are led by @scaling01, whose post about 'RT @elonmusk: Grok foundation ...' garnered the highest engagement. Key themes trending across the top stories include huawei, agents, every, human, their. The community is actively discussing recent developments in AI, engineering practices, and startup strategies.

1. scaling01 (Group Score: 498.1 | Individual: 54.5)

Cluster: 18 tweets | Engagement: 12444 (Avg: 274) | Type: Tech

RT @elonmusk: Grok foundation model V9-Medium (1.5T) has finished training. Evals look good. A lot of Cursor data was added in supplementary training and there is more to come.

Fine-tuning is underway and reinforcement learning begins in a few days. 2 to 3 weeks to public release.

This will be a major improvement over the 0.5T v8-small that currently serves all Grok production traffic, especially for difficult coding tasks.

See 17 related tweets

@eliebakouch: only 2-3 weeks for the RL phase of a 1.5T model is an interesting data point. also interesting that ...
@minchoi: Grok just turned Cursor data into a coding beast.

> 1.5T params > 3x the current Grok model &...

@testingcatalog: SPEACEXAI 🔥: The next Grok model is expected to be ready for public release in 2-3 weeks.

> 1.5...

@XFreeze: Elon Musk just gave a huge update on the next-generation Grok foundation model

The new V9-Medium mo...

@mark_k: 1.5T. Cursor. Grok. Music to the ears of AI people.

Grok V9-Medium from @xai has reportedly finishe...

2. lennysan (Group Score: 227.0 | Individual: 68.5)

Cluster: 6 tweets | Engagement: 1674 (Avg: 178) | Type: Tech

My biggest takeaways from @danshipper:

The future of work will happen inside Codex or Claude Code. Instead of putting AI into your SaaS tool, you’ll use your SaaS tools inside your favorite AI agents' in-app browser. Dan spends all his time in Codex now—writing documents, managing email, doing research, everything. He's using Google Docs, PostHog, and everything he needs within the agent's in-app browser. The agent can see what he’s doing, and has all of his context, so he and his agent collaborate quickly and super effectively.
Automation is a lie—every automation needs a human. Dan's company doubled in size this year despite being incredibly AI-forward. Why? Because in order to make automation work well, you need humans making sure everything keeps working. This is why benchmarks are misleading—they measure AI on problems we’ve already framed and can score, but there’s always a higher frame.
PMs will win the AI era. Marcus, a former PM who previously ran Axios’s writing product, joined Every after getting super AI-pilled. Now he runs their product Spiral, and ships faster than anyone on the team. He pairs technical knowledge with spiky product sense, deep user empathy, and an eye for what matters. Dan thinks any PM who gets really AI-native will be incredibly dangerous because the building is done for you—what matters is figuring out what to build and if it’s great.
Full-stack designers are becoming superheroes. Designers used to make beautiful interactions that engineers didn’t want to build or couldn’t execute properly. Now designers don’t need to hand things off; they can build it themselves. Designers are naturally creative people, and AI is the perfect tool for them because it lets them bring their vision to life without the traditional bottlenecks.
SaaS is not dead. In fact, Dan is bullish on SaaS stocks. When users bring their own AI (via Codex or Claude Code) to use SaaS products, the user—not the SaaS company—pays for tokens. This saves SaaS company’s margins. Since the agents need their own seats, Dan predicts that agents will create massive new demand for SaaS because there will be tons of agents using these products at high volume.
Every company will have one “super-agent” inside their Slack that every employee will use. Dan initially thought every employee would have their personal work agent, like a shadow AI org chart, but he’s completely flipped his view. He realized agents need humans who care about them. When someone gets tired of maintaining their personal agent, it becomes useless. The winning model is one forward-deployed engineer or AI-savvy person who maintains a company-wide agent (like Shopify’s River or Viktor), and then it trickles down to more specialized team agents as models improve and become less fiddly.
The AI job apocalypse is not happening, but you do need to evolve to stay relevant. Models make yesterday’s human competence cheap. But because everyone uses the same models, it all looks the same if you use it the default way; it becomes commoditized slop. Humans then take that frozen competence and use it to make something new and interesting for their specific situation. The key: “ride the models”—use them for everything you do, try new models when they drop, keep turning over rocks.
We will read way more AI-generated writing, and we will like it. Human writing is incredibly important for things that matter, but for internal docs, planning, and email, AI-generated is often better because most people are bad at writing strategy documents.
Build software for humans and agents to use together. The current model is building a CLI that an agent uses independently. Instead, you and your agent should be using the app together. This creates new design challenges—agents can make a billion requests in three seconds, so you need approval flows, inboxes that summarize what happened, logs, and easy rollback.
Forward-deployed engineers are the new most essential role. The big model companies have teams of people managing their internal agents, and those teams aren’t going away. It’s different from traditional software building, and certain engineers love it. As models get better, this role will evolve—you’ll be managing more agents doing more things.\n\nQT @lennysan: Automation is a lie. CLIs are over. The SaaSpocalypse is dumb.

A year ago @danshipper came on the podcast to predict where AI was heading. He was remarkably right—including the call that everyone was sleeping on Claude Code.

Dan has a unique lens into where things are going because his team at @every is possibly the most AI-pilled group of people in tech. I always learn a ton talking to Dan.

So I brought him back for round two. We'll score these in exactly a year: 🔸 Every company will have one “super-agent” in Slack. 🔸 Codex and Claude Code will become the new operating system for knowledge work. 🔸 The AI job apocalypse is not happening. 🔸 PMs and designers will thrive. 🔸 We will read way more AI-generated writing and we will like it. 🔸 "I would buy SaaS stocks right now."

Listen now 👇 https://t.co/wzxQ5bz49h

See 5 related tweets

@every: RT @lennysan: My biggest takeaways from @danshipper:

The future of work will happen inside Codex...

@pmarca: Interesting.\n\nQT @lennysan: My biggest takeaways from @danshipper:

The future of work will hap...

@lennysan: .@danshipper: "Automation is a lie. Every time you automate something, you need a human on top of it...
@danshipper: RT @realironnam: some ideas from @danshipper’s latest episode with @lennysan that are worth sitting ...
@TrungTPhan: RT @lennysan: Automation is a lie. CLIs are over. The SaaSpocalypse is dumb.

A year ago @danshipper...

3. lukOlejnik (Group Score: 213.5 | Individual: 36.8)

Cluster: 6 tweets | Engagement: 52 (Avg: 108) | Type: Tech

"social control made possible by the massive collection of data and use of algorithmic systems. When every action—movements, purchases, relationships and preferences—leaves a trace, a new form of power emerges, namely the power to profile, predict and influence behavior, often without individuals being fully aware of it"\n\nQT @lukOlejnik: The first encyclical of @pontifex Pope Leo XIV is a major intervention on AI, human dignity, truth, war, and technological power. It is a warning that technology is never neutral when it is shaped by money, control, secrecy, and force. It rejects surrendering human judgment to AI.

The encyclical treats cyberattacks as part of a wider transformation of conflict. Cyber conflict creates instability before formal war starts. War no longer is limited to tanks and missiles. It can begin with data theft, infrastructure disruption, manipulation, and invisible attacks whose authors are hard to prove.

The enciclica considers also information operations and cognitive warfare. It posits that propaganda, AI-generated manipulation, fear campaigns, and cyber operations are part of hybrid conflict. The battlefield includes imagination, trust, identity, and social cohesion. This is the cognitive dimension.

The encyclical addresses lethal autonomous weapons systems, and AI-assisted targeting. Here the message is clear: lethal decisions cannot be delegated to machines.

Pope Leo XIV’s first encyclical therefore cannot be reduced to AI only. It is about the kind of civilization being built around AI. https://t.co/WikmGzSjed

See 5 related tweets

@lukOlejnik: "Entire regions are currently subjected to a new mindset of extraction: that of health data, epidem...
@lukOlejnik: "Moreover, ownership of data cannot be left solely in private hands but must be appropriately regul...
@lukOlejnik: "Public opinion is gradually being shaped and conditioned by polarizing media narratives, which are ...
@lukOlejnik: "Today, among the goods that are universally intended for everyone, we must also include new forms o...
@lukOlejnik: Pope @Pontifex on cryptocurrencies: "finance has increased in importance and has undergone significa...

4. rohanpaul_ai (Group Score: 177.3 | Individual: 33.2)

Cluster: 7 tweets | Engagement: 18 (Avg: 50) | Type: Tech

🇨🇳 Huawei reveals a new chip design breakthrough under US sanctions pressure.

A design approach meant to close the gap with TSMC and Intel without relying only on smaller transistors, by making chip signals travel less distance.

They want 1.4nm-class density without owning the world’s best lithography tools i.e. they are trying to replace Moore’s Law with Tau Scaling Law.

To note, Huawei has been blocked from normal access to TSMC since the US tightened foreign direct product rules around Huawei in 2020, and TSMC later said it had not supplied Huawei since mid-September 2020.

Proposed τ Scaling as a new way to make chips faster when shrinking transistors is no longer delivering the same gains.

Said its next Kirin phone chip will be the first full test of Tau Scaling Law,

Old chip progress mostly came from making every transistor smaller, but Huawei’s idea shifts the target from smaller geometry to shorter signal delay, meaning less time wasted while electrical signals crawl through wires, gates, memory paths, and system links.

LogicFolding attacks the circuit layout itself by folding logic blocks closer together, shortening critical wires, reducing resistance and parasitic capacitance, and letting signals switch faster with denser placement.

So LogicFolding is the circuit-level piece: it tries to place related logic closer together, shorten key wires, cut electrical drag from resistance and parasitic capacitance, and raise performance without needing a full manufacturing-node leap.

Huawei is also pushing the same timing idea across the full stack: transistors, circuits, chip architecture, software scheduling, and system interconnects all get tuned to reduce τ, the delay constant that limits speed and efficiency.

The bold claim is that Huawei has already mass-produced 381 chips using this thinking, and future high-end chips could reach density comparable to 14Å, or 1.4nm, without relying only on classic process shrinkage.

Says this path could reach 1.4nm-class, or 14Å-class, density by 2031, while TSMC and Intel target similar physical nodes around 2029.

Huawei calls it Her’s Law, after He Tingbo, the chip leader who helped turn HiSilicon into Huawei’s survival engine after US export controls.

huawei. com/en/news/2026/5/ieee-iscas-tau-scaling\n\nQT @rohanpaul_ai: 🇨🇳 🇺🇸 China's Huawei’s new 122TB SSD shows how export controls can move innovation sideways instead of simply stopping it.

Huawei just built a 122.88TB AI SSD by changing the package around the memory, not by matching Samsung’s most advanced 400+ layer 3D NAND.

And a 245TB version discussed as a future step.

High-capacity SSDs usually grow by stacking more NAND layers inside each chip, but Huawei’s access to those chips is blocked because its Entity List status restricts items tied to US technology.

So it is not trying to win only by making taller 3D NAND stacks, where Samsung has already shown 400-plus-layer V-NAND work.

Instead, Huawei is shifting the contest from the chip itself to the way chips are packed together.

Huawei’s workaround is Die-on-Board, which puts NAND dies directly onto the circuit board, cuts out some normal chip packaging, and raises board-level density by packing more lower-density memory into the same device.

Direct die placement creates heat and signal problems, but it shows how packaging can recover some of the capacity lost when a company cannot buy the best memory chips.

See 6 related tweets

@WesRoth: Moore’s Law is getting harder, slower, and more expensive.

Huawei presented the Tau Scaling Law, a ...

@wallstengine: HUAWEI HAS A NEW PATH AROUND CHIP SCALING LIMITS

Huawei just unveiled its Tau Scaling Law, a design...

@StockSavvyShay: Huawei claims its new LogicFolding approach can help narrow the gap with $TSM by improving chip dens...
@scaling01: the last race of our time

(a player with enough compute can still catch up to OpenAI and Anthropic....

@Cointelegraph: 🇨🇳 JUST IN: Huawei claims a chipmaking breakthrough that could close its 5-year gap with TSMC, targe...

5. WesRoth (Group Score: 171.4 | Individual: 39.6)

Cluster: 5 tweets | Engagement: 106 (Avg: 29) | Type: Tech

This is the AI pricing problem people have been quietly avoiding.

For the last two years, a lot of AI tools felt artificially cheap because companies were subsidizing usage to win users, train habits, and grab market share.

That phase may be ending. https://t.co/IgjkHsjH5q\n\nQT @HedgieMarkets: 🦔Microsoft canceled its internal Claude Code licenses this week after token-based billing made the cost untenable, even for a company with effectively infinite cloud resources. Uber's CTO sent an internal memo warning the company burned through its entire 2026 AI budget in just four months. American AI software prices have jumped 20% to 37%, and GitHub (owned by Microsoft) is dropping flat-rate plans for usage-based billing across its products.

My Take The AI subsidy era is ending in real time. The same company that put $13 billion into OpenAI and built the Azure infrastructure powering most of Anthropic's compute just looked at the bill from a competitor's coding tool and decided it was not worth paying. That is not a productivity failure on Anthropic's end. Token-based pricing is forcing every enterprise customer to confront the actual cost of running these models at scale, and the number turns out to be far higher than the flat-rate experiments suggested.

This ties directly to my Gemini Flash post yesterday. Anthropic, OpenAI, and Google all raised effective prices in the last six months. Enterprises that built workflows assuming AI costs would keep falling are now watching annual budgets evaporate in months. Two outcomes look likely from here. Either enterprises scale back AI usage to fit budgets, which slows the revenue ramp the labs need to justify their valuations ahead of IPOs, or the labs cut prices and absorb the losses, which makes the unit economics worse at exactly the wrong moment. Both paths land in the same place, the numbers stop working, and somebody has to take the writedown.

Hedgie🤗

See 4 related tweets

@alex_prompter: RT @alex_prompter: Marc Andreessen admitted on Joe Rogan that AI is making people less efficient.

T...

@aakashgupta: The framing here is exactly backwards. This is the strongest AI bull signal anyone has published thi...
@VaibhavSisinty: In the last 30 days, Microsoft canceled Claude Code for thousands of engineers, Uber burned its enti...
@zerohedge: RT @cryptorover: 🚨 THE AI COST CRISIS HAS STARTED.

Microsoft reportedly told engineers to stop usin...

6. _chenglou (Group Score: 166.5 | Individual: 27.3)

Cluster: 10 tweets | Engagement: 7438 (Avg: 2625) | Type: Tech

https://t.co/FmCx6Ku6Ll\n\nQT @disclosetv: NOW - Pope XIV says the church and Anthropic, will work together to "find the way for humanity, in this time of artificial intelligence." https://t.co/b1BB0TeQep

See 9 related tweets

@etnshow: BREAKING: Vatican announces Anthropic partnership https://t.co/Rtlq5MNzz4\n\nQT @disclosetv: NOW - P...
@godofprompt: This wasn’t on my bingo card\n\nQT @disclosetv: NOW - Pope XIV says the church and Anthropic, will w...
@Hesamation: Anthropic Holy Trinity https://t.co/WTLoipjE98\n\nQT @disclosetv: NOW - Pope XIV says the church and...
@tunguz: Pope Leo XIV just released his first encyclical, Magnifica Humanitas, on the impact of Artificial In...
@irl_danB: life immitates parody https://t.co/da3dAvlCJR\n\nQT @disclosetv: NOW - Pope XIV says the church and ...

7. rohanpaul_ai (Group Score: 139.9 | Individual: 32.2)

Cluster: 5 tweets | Engagement: 31 (Avg: 50) | Type: Tech

Grok Build beta (agentic coding agent that lives in your terminal) is now open to all SuperGrok and X Premium+ users.

Before only SuperGrok Heavy users had access.

You install it once with a single command, then you can type natural-language instructions like: “Make a rollercoaster simulator” or “Build a todo app in the style of Apple” etc..

and Grok will:

Plan Mode -> create a step-by-step plan you review and approve before it touches any code
Use sub-agents that run in parallel for complex tasks
Edit multiple files, use git, run tests, search the web, etc.
Create images and videos using Imagine (Grok’s image/video generator) right inside the workflow
Build automations or full orchestrators (you can turn sessions into reusable “skills” or connect multiple agents)\n\nQT @xai: Grok Build is now available in Beta for all SuperGrok and X Premium+ users.

Use Plan Mode, create images and videos with Imagine, and build automations or orchestrators with the CLI.

Visit https://t.co/bpTHpjivWD to get started. https://t.co/OZ0kjtkpUf

See 4 related tweets

@XFreeze: Grok Build is now officially available to both SuperGrok and 𝕏 Premium+ users

You should seriously ...

@BrianRoemmele: Wow! This is one of the most powerful tools from a large platform brought to the lower cost Premium+...
@WesRoth: Grok Build is reportedly being prepared for rollout to SuperGrok users, expanding access beyond the ...
@yagiznizipli: RT @myrhex: Grok Build is now officially available to all SuperGrok users.

You can install it rig...

8. alex (Group Score: 134.1 | Individual: 48.5)

Cluster: 5 tweets | Engagement: 3952 (Avg: 412) | Type: Tech

RT @AP: Pope Leo XIV called for robust regulation of artificial intelligence and for its developers to work for the common good rather than profit, issuing a sweeping manifesto on safeguarding humankind as the technology impacts everything from work to war. https://t.co/VSlFOkxhry

See 4 related tweets

@FT: Pope Leo has called for AI to be 'disarmed' through robust regulation and limits on its use for warf...
@ProudSocialist: Pope Leo XIV has just released a historic manifesto on artificial intelligence, calling it the bigge...
@FirstSquawk: Pope Leo XIV warned that artificial intelligence “threatens to normalize an anti-human vision” and s...
@techdaily24: Pope Leo XIV says AI must serve humanity, not the powerful few https://t.co/4LYyA4FDWc #startups #te...

9. Scobleizer (Group Score: 111.6 | Individual: 44.8)

Cluster: 4 tweets | Engagement: 652 (Avg: 97) | Type: Tech

RT @DavidSacks: The Pope rightly warns that AI must serve human dignity, not become a tool of domination or exclusion.

But if we hand governments sweeping power over AI development in the name of safety, how do we prevent it from being used to censor, surveil, and control citizens — as Orwell foretold in 1984?

This is the real alignment problem.

“Quis custodiet ipsos custodes.” Who will guard the guardians?

“Power tends to corrupt, and absolute power corrupts absolutely.”

The oldest questions of human nature and authority don’t disappear in the AI age. They become newly relevant.

See 3 related tweets

@MarioNawfal: 🇻🇦 The Vatican and Anthropic have joined forces to protect humanity in the age of AI.

Pope Leo is s...

@ProudSocialist: In his landmark manifesto, Magnifica Humanitas, Pope Leo XIV calls for AI to be disarmed:

“Artific...

@reach_vb: The Pope said AI should keep humans at the centre

Codex: handling the boring parts so humans can re...

10. photomatt (Group Score: 104.8 | Individual: 41.2)

Cluster: 3 tweets | Engagement: 1414 (Avg: 328) | Type: Tech

RT @itsolelehmann: the pope and anthropic's co-founder just stood together at the vatican to release "magnifica humanitas," the first ever catholic teaching on AI

yes, you read that right. the full ceremony was 2 hours.

here's the most interesting things for you to know:

this is the biggest religious response to AI in history. popes only put out a handful of these huge official letters in their entire time as pope. the fact that one of them is about AI tells you how seriously the church is taking what's coming.
small detail with massive meaning: this pope picked the name "leo XIV" on purpose. the last pope named leo was leo XIII back in 1891, and his most famous act was writing the church's response to the industrial revolution. picking the same name is a deliberate signal. this pope sees AI as the new industrial revolution.
the catholic church does this every time a major technology reshapes humanity. they wrote "rerum novarum" in 1891 to respond to the industrial revolution. when nuclear weapons threatened the world in the 1960s, they wrote "pacem in terris." climate change and runaway tech got "laudato si" in 2015. now AI gets "magnifica humanitas." they don't issue these often.
the pope's main line: "AI needs to be disarmed." he literally compared AI to nuclear weapons. he said the church spent decades pushing for nuclear disarmament because the technology was too dangerous to leave in the hands of a few. he says AI is now in that same category.
anthropic co-founder christopher olah told the pope, on stage at the vatican, that anthropic's own research team keeps finding things inside their AI models that "mirror joy, satisfaction, fear, grief, and unease."
olah's reframe of what AI actually is: these things are grown. they're trained on a structure roughly modeled after the human brain and fed everything humans have ever written. in his own words: "they are made from us, from our words." he said even the people building them don't fully understand what's happening inside.
olah publicly admitted that every AI lab, including his own, faces pressure that can conflict with doing the right thing. commercial pressure to keep shipping, competitive pressure from other labs, plus the older pressures of pride and ambition. his solution: we desperately need outside critics with no skin in the game who will tell the labs when they're failing.
olah says there are 3 giant questions the AI labs cannot answer alone and the world needs religion and philosophy to step in on:

how do we make sure poor countries actually benefit from AI? what does human flourishing even look like in this new world? and what are these things we're actually building?

one of the sharpest lines in the whole encyclical: "the promise of automatic general prosperity often proves illusory." translation: the idea that AI will just make everyone rich on its own is a fantasy. someone has to actually design the system so the benefits get shared.
the pope also pulled out a 100-year-old quote: "contemporary man has not been trained to use power well." said by a theologian back in the 1920s. the whole encyclical is basically a long argument that we need to learn how to use this kind of power before it uses us.
the pope kept stressing that he doesn't have the technical answers. but he says the church has thousands of years of wisdom on what it means to be human, and that wisdom is exactly what's missing from how we're building AI right now. his closing line: this technology should serve "human flourishing and human dignity, not control consciences."

See 2 related tweets

@AISafetyMemes: The Pope issued an urgent warning to "slow down" AI before it "dominates humanity"

WHY THIS IS UNUS...

@pfau: Kind of embarrassing watching the pope angling for an MTS gig at Anthropic like this.\n\nQT @itsolel...

11. aakashgupta (Group Score: 99.5 | Individual: 34.1)

Cluster: 3 tweets | Engagement: 96 (Avg: 173) | Type: Tech

Everyone keeps asking if AI is going to kill the PM role.

My guest this week runs an AI-native company, and the answer was the opposite of the hype. They're hiring more PMs than ever. More engineers than ever.

But the bar moved.

Code is cheap to produce now. So the PMs who stand out aren't the ones who write the cleanest spec. They're the ones with a real opinion about what to build. Product taste is the alpha.

So I asked the obvious follow-up. Where does product taste actually come from?

The answer was more concrete than I expected.

Taste is an output. It comes from consuming an enormous volume of feedback from every direction at once.

The best PMs already do this manually. YC has been telling every cohort the same two words for years: talk to users. But "talk to users" in 2026 means something bigger than customer calls.

It means building a context graph. One connected feedback layer that pulls from:

Wherever your team logs issues
GitHub discussions
Slack and Discord conversations with your community
Gong transcripts from every customer call
Product analytics from PostHog, Amplitude, Pendo, and FullStory
Twitter, if your users post about you there

Every one of those is a feedback stream. Most teams let them sit in separate tools, read by separate people, never connected.

Here's the part that reframes the whole thing.

Once that feedback lives in one graph, a human doesn't have to be the one reading it. Your agent can consume it.

That changes what a PM's day looks like. Instead of manually synthesizing scattered signal, an agent reads the full context graph and surfaces the patterns. You spend your time on judgment, not collection.

This is why I don't buy the "death of the PM" narrative.

The PMs who lose are the ones who thought the job was writing requirements. The PMs who win are the ones building the feedback machine that taste is actually made of.

Taste used to feel innate. It's becoming a system you can engineer.\n\nQT @aakashgupta: She literally broke down how to run evals in Claude Code (built the whole thing live):

01:34 - What people get wrong with evals 04:35 - Why product taste is the alpha now 09:28 - Building a PM agent from one prompt 19:00 - Instrumentation without writing code 22:00 - Watching traces stream in live 28:00 - Getting Claude to write your first eval 33:58 - When vibe evals work and when they don't 48:50 - The self-improving loop (this part is wild) 01:03:00 - Same-day shipping is real 01:06:00 - The context graph unlock

See 2 related tweets

@aakashgupta: The reason 99% of AI agents ship without evals has nothing to do with technical complexity.

The act...

@aakashgupta: On this week's episode, my guest built a working PM agent live, on camera.

Here's what it did.

It ...

12. DynamicWebPaige (Group Score: 97.9 | Individual: 41.0)

Cluster: 3 tweets | Engagement: 396 (Avg: 76) | Type: Tech

RT @OfficialLoganK: We just launched the ability to build native Android apps directly in Google AI Studio for free!

Since launch last week, people have created more than 250,000 Android apps. Likely >99% of these folks never built an Android app before, everyone can now build, no coding required! https://t.co/3pNyVMfg56

See 2 related tweets

@startupideaspod: Android has billions of active users.

You can now build native apps for them in AI Studio.

No code...

@nummanali: Talk about eating someone’s lunch

Google AI Studio lets you build native Android apps now\n\nQT @O...

13. mollytaft (Group Score: 93.8 | Individual: 58.8)

Cluster: 3 tweets | Engagement: 9308 (Avg: 759) | Type: Tech

RT @ChristopherHale: In leaked audio recordings, Peter Thiel revealed he told JD Vance to ignore Pope Leo XIV on moral issues, including the development of ethical AI.

Calling Leo XIV the “woke American pope,” he also suggested the American pope was a tool of the Antichrist.

Now, Pope Leo XIV is releasing what’s expected to be an historic encyclical addressing AI tomorrow morning. https://t.co/i3RoHOhJQM

See 2 related tweets

@Reuters: 🔊 'This is a key concern for him, for his papacy, for the Vatican.' @joshjmcelwee on why Pope Leo is...
@MyLordBebo: 🤖🇻🇦 🇺🇸 Pope Leo XIV will collaborate with Anthropic, American AI safety & research company to help i...

14. BrianRoemmele (Group Score: 93.0 | Individual: 33.0)

Cluster: 3 tweets | Engagement: 118 (Avg: 527) | Type: Tech

Anthropic’s Appeal To The Pope Has An Echo In The Past

The person this finger belonged to was arrested, his OPEN work was censored because IT WAS NOT SAFE.

Today “experts” beg for censorship.

This is the middle finger of Galileo Galilei and is a message for Anthropic… https://t.co/zwHoRnlLES\n\nQT @BrianRoemmele: Talking To The Pope: Anthropic’s Latest Interpretability Claims: AI Regulatory Capture

Gatekeeping in Action: Fear and “Safety” as Competitive Moat and Regulatory Lever

In a presentation alongside Pope Leo XIV at the launch of the encyclical Magnifica Humanitas, Anthropic co-founder Chris Olah highlighted “mysterious and unsettling” discoveries in AI models.

He described internal structures that mirror human neuroscience findings, evidence of introspection, and functional internal states resembling emotions such as joy, satisfaction, fear, grief, and unease.

Olah admitted uncertainty about their meaning but called for “ongoing discernment.”

This narrative, drawn from Anthropic’s interpretability research (including papers on emotion concepts in Claude Sonnet 4.5 and introspective capabilities in Opus 4 models), serves a dual purpose: it generates awe and concern while reinforcing the company’s preferred approach to AI development.

Far from neutral scientific observation, these claims fit into a broader pattern where Anthropic uses selective openness, safety rhetoric, and policy influence to gatekeep advanced AI capabilities for a privileged few: incumbents with the resources to navigate (and shape) the resulting regulatory landscape.

Rebuttal to Olah’s Claims in the Video

Claim 1: Structures that mirror results from human neuroscience. Anthropic’s work, building on earlier efforts like feature visualization and circuit analysis, identifies neuron activations and representations that parallel biological findings—e.g., abstract concept encodings or hierarchical processing.

Rebuttal: These parallels are unsurprising and overstated. Large language models are trained on vast corpora of human-generated text and data, which inherently encode patterns from human cognition, neuroscience literature, and cultural descriptions of the brain. Statistical optimization in transformers naturally produces efficient, compressed representations that resemble biological efficiency (e.g., sparse coding or hierarchical abstraction) without implying deeper equivalence or mystery. Similar “mirrors” appear in open-source models and earlier architectures; they reflect convergent evolution in information processing, not emergent souls or unpredictable agency. Treating them as profound justifies restricted research access rather than inviting wider scrutiny that could falsify or refine them faster.

Claim 2: Evidence of introspection. Recent Anthropic papers demonstrate models like Claude Opus 4 showing functional awareness of their own internal states distinguishing injected “thoughts,” referencing prior intentions, or modulating activations when instructed to “think about” concepts. This is presented as early signs of meta-cognition.

Rebuttal: This is sophisticated pattern-matching and activation steering, not genuine introspection or self-awareness. Models are predicting what an “introspective” assistant persona would output or do, based on training data full of human self-reflection examples. Experiments show unreliability and heavy context-dependence; performance drops outside narrow setups. True introspection implies subjective experience or robust self-modeling independent of prompts absent here. Anthropic’s own caveats note it is “highly unreliable.” Framing steerable activations as “introspection” anthropomorphizes the system to heighten perceived stakes, supporting arguments that only highly controlled, “responsible” labs should advance these capabilities.

1 of 2

See 2 related tweets

@BrianRoemmele: My public insight here makes me no friends in the AI (grow rich fast) crew.

You know, I don’t care....

@BrianRoemmele: RT @BrianRoemmele: Talking To The Pope: Anthropic’s Latest Interpretability Claims: AI Regulatory Ca...

15. Reuters (Group Score: 91.4 | Individual: 23.2)

Cluster: 5 tweets | Engagement: 297 (Avg: 113) | Type: Tech

Pope Leo warned that some autonomous weapons systems are 'practically beyond any human reach' to control, as he presented his first encyclical on the risks of AI systems https://t.co/89tIpLaQ8v https://t.co/xkuzoCwgSi

See 4 related tweets

@Reuters: Pope, urging AI regulation, warns some weapons now beyond human control https://t.co/aalqTA2Gsz http...
@Reuters: 🔊 'One of the things that Leo has been very vocal about is the use of AI in warfare.' Pope Leo will ...
@MyLordBebo: 🇻🇦 Pope Leo XIV warns that AI fuels modern warfare in first theological document — "We cannot consid...
@CNBC: Pope Leo urges AI regulation and warns that some weapons are now beyond human control https://t.co/W...

16. dair_ai (Group Score: 86.5 | Individual: 33.5)

Cluster: 3 tweets | Engagement: 69 (Avg: 71) | Type: Tech

New paper from Microsoft on Self-Evolving Agent Skills\n\nQT @omarsar0: New research from Microsoft Research

I see a lot of AI engineers handwriting agent skill docs and hope they generalize.

Probably not optimal. This works show why.

It treats the skill doc as a trainable external state of a frozen agent instead.

It introduces SkillOpt, where an optimizer model makes validation-gated edits to the skill file. It adds, deletes, or replaces instructions, with a textual learning rate that controls how aggressively each round rewrites the doc. The agent itself never changes.

SkillOpt is best or tied on all 52 (model, benchmark, harness) cells.

On GPT-5.5 it adds 23.5 points in direct chat, 24.8 with Codex, and 19.1 with Claude Code over no skill. It beats human-written skills, TextGrad, GEPA, and EvoSkill, carries zero extra inference-time cost, and the learned skills transfer across models and harnesses.

Paper: https://t.co/mNgTmmT32U

Learn to build effective AI agents in our academy: https://t.co/1e8RZKs4uX

See 2 related tweets

@Shashikant86: Microsoft released the SkillOpt research paper! It looks similar to SuperOpt and paper and CodexOpt ...
@kazunori_279: RT @omarsar0: New research from Microsoft Research

I see a lot of AI engineers handwriting agent sk...

17. chandrarsrikant (Group Score: 83.6 | Individual: 24.8)

Cluster: 5 tweets | Engagement: 131 (Avg: 2007) | Type: Tech

🚨IMP: IT ministry to look into Pronto's in-home AI recording pilot amid privacy concerns

The Ministry of Electronics and Information Technology has taken cognisance of the controversy around Pronto's in-home recording pilot and is looking into the matter, government sources told Moneycontrol, bringing greater regulatory scrutiny to how startups use customer-home data for AI systems.

The development comes after Bengaluru-based home-services startup Pronto confirmed it was testing an opt-in feature involving recordings during cleaning and other household tasks, triggering debate around surveillance, consent and the use of customer-home data for AI training purposes.

By @AryamanGupta_ and @AihikS

https://t.co/mOZXpSU4TJ

See 4 related tweets

@chandrarsrikant: In response to Moneycontrol's queries, Snabbit said it had only explored the broader space through v...
@moneycontrolcom: #Startups | Pronto’s in-home recording pilot raises privacy concerns for home-services startups

Sou...

@shaw_reshab: IT ministry to look into Pronto's in-home AI recording pilot amid privacy concerns

✍️ @AihikS &...

@tsuvik: IT ministry to look into Pronto's in-home AI recording pilot amid privacy concerns

https://t.co/h7s...

18. elonmusk (Group Score: 83.2 | Individual: 30.5)

Cluster: 3 tweets | Engagement: 26406 (Avg: 28659) | Type: Tech

Grok Build improving every day, 7 days a week\n\nQT @skcd42: Bug fixes shipping to Grok Build 0.1.219 (release notes will be available in the TUI)

fixing usage limit bugs with prompt caching
fix layout-shifted chars in kitty terminal
replace Shift+Enter w/ Alt-Enter on VTE-based terminals (GNOME Terminal, Ptyxis, kgx, Tilix, etc.)
default expand (search_replace, write_file, run_terminal_command) calls on scrollback
mixed prose + URL paste no longer silently drops prose
set_images byte: length collision losing chips
make multi-line markdown link URLs clickable across word-wrap
clear selection when expanding truncated group
remove read_file duplicate-read check that breaks after compaction

See 2 related tweets

@XFreeze: Grok Build just got another update

v0.1.219 is now rolling out with tons of new changes, release no...

@XFreeze: Grok Build just got another update:

v0.1.220-alpha.1

xAI is now shipping multiple updates in a sin...

19. romanhelmetguy (Group Score: 80.4 | Individual: 44.2)

Cluster: 2 tweets | Engagement: 17654 (Avg: 2137) | Type: Tech

Gonna try to explain this to tech CEOs again: Young Americans are pissed. They feel betrayed. Half have embraced the far right & want to cut off your access to cheap foreign labor. The other half have embraced the far left & want to cut off your head. One side will win. Choose.

See 1 related tweets

@pmarca: Concerning.\n\nQT @romanhelmetguy: Gonna try to explain this to tech CEOs again: Young Americans are...

20. adamlyttleapps (Group Score: 80.1 | Individual: 44.8)

Cluster: 2 tweets | Engagement: 234 (Avg: 65) | Type: Tech

Finally making some progress

1.84x ROAS

(For every dollar spent on marketing today I got $1.84)

Key learnings so far:

—

Increased prices to $49.99/yr or$ 14.99/m

With better margins you can afford $24.85 per acquisition. You can survive metas learning phase!

ASO pricing doesn’t work. You can’t afford weekly pricing and wait for them to stay subscribed for weeks. You need revenue quick (cashflow)

—

Created a web2app funnel (built my own)

App Store traffic is EXPENSIVE. And downloads are binary (either they download or they don’t.. no in between)

Plus: you gotta install all the arbitration stuff in your app. I really can’t be bothered.

Instead I’ve been building a app2web flow

Tried replicating cal ai and simply piano 32+ step flow but most people never got to the paywall

Instead I stripped it down to 3 screens: Hero, testimonial, paywall

Then A/B tested these screens endlessly until I started to get conversions

Spoiler: today it’s converting (finally)

Feed the meta pixel as much data as possible

My funnel triggers data each step of the way. This helps meta find the right users who will progress through the funnel.

Hardest part seems to be training the algorithm for the right users who purchase. But you need people purchasing to train. Catch 22

Trickle feeding the algorithm each step of the way seems to help meta nudge in the right direction.\n\nQT @adamlyttleapps: Scrapped TikTok ads

Focused on meta ads

Switched from turning intent from download → purchase

Got a trial signup within hours

Coincidence? https://t.co/UJSYuBkF1I

See 1 related tweets

@adamlyttleapps: It costs a lot to pay for app install traffic form meta and you gotta add attribution to your app fo...