gª|¡Þ€ªµ 高德威

e/acc | Co-Founder, SDGˣ | Director, Near Future Lab | Complexity Theorist | Forbidden Research @CMUhcii×@MIT | Researching Cyber-identity Disorder | @AIT | @UN

[gª|¡Þ€ªµ 高德威 @galipeau]

From Here to the Final Frontier: Understanding Star Trek's Post-Scarcity Economy and How We Might Actually Build One
https://t.co/znSWYAfvHS

[Camus @newstart_2024]

Tim Dillon had a brutally funny take on Gen Z’s approach to work.

He says a lot of them have figured out the whole system feels like a scam, so they’re treating it like one. Fake mental health days, quiet quitting, weaponizing HR language, doing the bare minimum while demanding maximum accommodation.

And Tim’s reaction? “I’m for it.” They’re just using the playbook society handed them.

This is what happens when trust in institutions and old-school work ethic collapses. People stop playing the game seriously and start playing the system instead.

Do you think Gen Z is smart for gaming a broken system, or is this approach ultimately making things worse?

[gª|¡Þ€ªµ 高德威 @galipeau]

@Catpostx Soon!!

[Trung Phan @TrungTPhan]

BREAKING: Audemars Piguet has partnered with McDonald's on a limited-edition chicken nugget. The iconic Swiss watchmaker will bring its legendary precision and 150 years of heritage to create world's most luxurious piece of shaped fried breaded white protein.

$499 per nugget.

[DaVinci @BiancoDavinci]

An instrument that sounds like human voices.

[Bangkok Post @BangkokPostNews]

Thailand has launched a government-backed Thai-language artificial intelligence project aimed at building a national large language model that understands local language, culture and context. The 80 million baht initiative is designed to strengthen digital infrastructure, support the tech sector and boost skills across the workforce. Officials say the project will help reduce reliance on foreign platforms and support businesses, developers and public agencies as Thailand expands its digital economy.

Listen to the story or get the full story in the 1st comment.

#BangkokPost #Thailand #AI #Technology

[Pichapen @pichapen]

If you are building in Bangkok, this event is for you 🫵

Following our success with "Thailand in the AI Race," we are taking the next step: Physical Infrastructure.

We are proposing that Tobacco Factory 5 (โรงงานยาสูบ 5) - the historic space right behind QSNCC and next to Benjakitti Park be transformed into the city's first AI Builder Hub.

By utilizing vacant BMA assets, we create a high-density zone where policy meets product. The "trust-layer" of AI including capital, talent, and infrastructure, which needs a center of gravity. This is it.

Join us at the site on May 12 for our AI DEMO DAY

📍 Location: Tobacco Factory 5, Benjakitti Park

The Agenda:
14:30 | Doors Open: Check-in via AI Passport
15:10 | Forum 1: Built in Bangkok -Why Builders Choose This City
16:00 | BKK Showcase: part 1
16:40 | Forum 2: Scaling the Engine - Capital, Infra & BKK’s AI Future
17:20 | The Hero Workshop: Live Build
18:20 | BKK Showcase: part 2

Bangkok is not staying on the sideline, Bangkok is the platform.

Registration link below 👇

[Milk Road AI @MilkRoadAI]

This is one of the craziest AI launches of 2026 and it came out of basically nowhere (Save this).

A company called Subquadratic just shipped SubQ, and the benchmarks are almost hard to believe.

To understand why this is such a big deal, you have to understand the fundamental problem that has defined AI for the last decade.

Every large language model in existence is built on transformer architecture, and transformers use a mechanism called standard attention that checks every single word in a sequence against every other word.

Double the context length and compute doesn't double, it quadruples, triple it and compute goes up nine times.

This quadratic scaling is why frontier models have been stuck at roughly 1 million tokens, why running them at those lengths gets expensive fast, and why the AI labs have essentially been printing money charging you more the longer you need the model to think.

The industry has known this problem existed since 2017 but they scaled it anyway. SubQ is built from the ground up to solve it.

Instead of processing every possible token relationship, SubQ's sparse attention architecture identifies which relationships actually matter and ignores the rest meaning compute is used where it counts and wasted nowhere else.

The result is that compute scales linearly with context length instead of exponentially, and the implications of that one architectural shift are enormous.

At 12 million tokens, SubQ reduces attention compute by nearly 1,000x compared to standard frontier models and at 1 million tokens, it runs 52x faster than FlashAttention.

And it does all of this while posting frontier level accuracy, scoring 95% on the RULER 128K long-context benchmark versus Claude Opus 4.6's 94.8%, and an 81.8 on SWE-Bench Verified coding tasks, besting Opus 4.6 (80.8) and DeepSeek 4.0 Pro.

The cost comparison is where it gets genuinely insane.

SubQ runs at under $1.50 per million tokens less than 5% of what Claude Opus charges.

On the RULER benchmark, running the test with SubQ cost $8, running the same test with Claude Opus cost $2,600 and that's a 300x cost reduction at equivalent or better accuracy..

Subquadratic launched with $29 million in funding, SubQ is available today for early access via API, and SubQ Code, a coding agent built on the architecture ships alongside it.

The transformer has been the unchallenged foundation of every major AI system since 2017.

SubQ is the first serious evidence that something structurally better might have just arrived.

[Bangkok Post @BangkokPostNews]

True Corporation has partnered with Google to launch the “AI for All Thais” programme, aimed at expanding access to artificial intelligence education across Thailand. The multi‑phase initiative, backed by the Ministry of Higher Education and universities nationwide, seeks to equip Thai students and workers with future‑ready digital skills. Courses will range from basic AI knowledge to a credit‑bearing curriculum for universities, as Thailand addresses a shortage of skilled AI professionals and pushes to strengthen long‑term competitiveness.

Listen to the story or get the full story in the 1st comment.

#BangkokPost #AI #education #Thailand

[gª|¡Þ€ªµ 高德威 @galipeau]

27862 30

[Jewel 🌗 @OfficialJoel4_]

ANOTHER BIG CHALLENGE

Nobody can get it correctly….

What Numbers Do you see 👀? RT

Correct deserve $600

Ends 48 hours

[Carlos E. Perez @IntuitMachine]

Meet Human Operator from MIT Media Lab: a wearable that lets AI temporarily take control of your hand using electrical muscle stimulation.

Watch it crush piano, draw perfectly, and mix cocktails like a pro — all from a simple voice command.

“I gave an AI a body.”

This isn’t sci-fi. This is tomorrow.

#HumanOperator #MITMediaLab

[Tsinghua University @Tsinghua_Uni]

Hats off to Innovative Tsinghua’s Assist. Prof. Bin Gui of the Yau Mathematical Sciences Center, for resolving a longstanding conjecture in rational conformal field theory, published in Inventiones Mathematicae, one of the top maths journals.

[Bryan Johnson @bryan_johnson]

To predict the next dominant societal philosophy, just measure the depth of the technological disruption preceding it. Every major shift breaks something and then ideology rushes in to patch what was lost.

For example, the printing press broke the Church's monopoly on truth and Protestantism patched it. The Industrial Revolution broke guild identity and Marxism and nationalism patched that.

AI is breaking more than we have words for yet. Nothing out there is big enough to patch what is breaking. We'll need something new.

[gª|¡Þ€ªµ 高德威 @galipeau]

Ah, the exquisite art of intellectual gullibility ... where one mistakes a well-polished non sequitur for profound epistemology, all while clutching a confirmation bias like a security blanket from Plato's cave.

[Sawyer Merritt @SawyerMerritt]

Bill Maher said tonight that Elon Musk has been the smartest on the subject of AI.

"I am very close to the cutting edge in AI and it scares the hell out of me. By the time we are reactive with AI regulation it is too late. AI is a fundamental existential risk for human civilization, and I don't think people fully appreciate that." - Elon Musk in 2017/2018

[Isaak @isaakfreeman]

I’m leaving MIT and not continuing into my PhD. AI is coming too fast for humans to keep up.

But there might be a way: I realized digital humans are more possible than most think. With capable AI researchers helping, maybe for $10B, maybe in less than 10 years, on 50k H100s.

[The Economist @TheEconomist]

Kidfluencing is a lucrative family profit centre, as content is spun out of everything from potty training to puberty. A new book looks at a business where “children’s privacy is traded for profits” https://t.co/QRZJVVO7XR

[Brian Halligan @bhalligan]

🚨🚨NEW EPISODE DROP  Sally Kornbluth, President of MIT🚨🚨

"If you take a lick of the lollipop of mediocrity, you will suck forever."

I love this.

She was talking about how MIT sustains excellence after 150+ years. And it applies to founders and CEOs who are trying to scale, more than ever.

The thing is, the bar doesn't drop all at once. It drops one hire at a time, one exception at a time.

It's better to have nobody than to let the standard slip even once. Every hire, every door, every message has to carry the same signal. That's how MIT maintains it's intensity.....just look at the list of grads like @vkhosla @gdb @drewhouston @mansourtarek_ @mntruell @_mohansolo @ChaseLochmiller

I’ve watched this play out with the companies I work with. They hit 150 employees, the founder stops interviewing everyone, and by 500 they're asking what happened to the culture.

Sally's answer: it's a lot easier to stop that slide than to recover from it. The fix is relentless consistency, clarity, and the willingness to say no far more often than you say yes.

Worth remembering when growth feels urgent.

------

Listen to the full conversation with MIT President Sally Kornbluth on Long Strange Trip.👇

[The Free Press @TheFP]

Anthropic’s latest model, Mythos, is stirring debate about whether AI must be placed under a Manhattan Project-style federal authority, writes Josh Code. https://t.co/GCoZ43jdqf

[Klaus Schwab @ProfKlausSchwab]

Universities Must Reinvent Themselves for the Intelligent Age

My latest essay in @TIME
➡️🔗 https://t.co/ItLpK9uvp7 #IntelligentAge

[Reuters @Reuters]

Anthropic talking to the Trump administration about its next AI model, co-founder says https://t.co/M5Yr15eKY3 https://t.co/M5Yr15eKY3

[Futurism @futurism]

Geniuses at work. https://t.co/SllcPXCtAg