theseouldan.bsky.social
@theseouldan.bsky.social
Pinned
In the era of AI crap, useless google search and cut and paste media, it is hard but I do think when you come across a surprising claim you should analyse the source of that claim
Reposted
If society doesn't normalize instantly knocking stalker glasses off people's faces, then I guess the alternative is to simply mask up all the time
Meta Plans to Add Facial Recognition Technology to Its Smart Glasses
www.nytimes.com
February 13, 2026 at 3:50 PM
Need to renormalise violence against people who wear these
Wow: Meta has been working on plans to add facial recognition technology to its AI smart glasses. nyti.ms/3Os1oxf

And this was the company’s cynical view on when, and how, to do it:
February 13, 2026 at 2:53 PM
Until the next time grandpa Simpson comes up with another number
The agreement ends a disadvantage for Taiwan as the new US tariff of 15% puts it on a par with America's other important trading partners, like Japan and South Korea.

By @aimewilliams.bsky.social

www.ft.com/content/c7e1...
US and Taiwan sign trade agreement to seal chip investment
Deal will reduce American tariffs on a range of Taiwanese food products
www.ft.com
February 13, 2026 at 5:06 AM
Reposted
February 10, 2026 at 5:51 AM
It should be a criminal offence to use this product and complain when you lose all your money
Coinbase rolls out Agentic Wallets, a wallet infrastructure that lets AI agents make autonomous crypto transactions "without human intervention" (Daniel Kuhn/The Block)

Main Link | Techmeme Permalink
February 12, 2026 at 4:21 AM
Fyi, if you see LLM generated "research" then they still can't add
I had to rethink a lot of priors during the 6 months last year when LLMs went from “can’t add” to “can’t do undergrad hw math” to “can’t find group isomorphisms it takes me an hour to do by hand” to “Can’t reinvent obscure parts of my thesis from scratch reliably given only a vague description”
Unpopular opinion here (and I’m not even an AI booster, I’m mostly a skeptic), but I think a whole lot of people are overindexed on AI being the too-many-fingers-and-weird-teeth machine and aren’t accepting or even understanding that its output has generally been continuously improving.
February 12, 2026 at 2:46 AM
Binary options were a popular scam about a decade ago
JFC this shit is out of control. 5 minute bets on crypto. These people belong in jail.
February 12, 2026 at 2:43 AM
Surely the people most at risk from LLMs are the people pumping out this sort of worthless "research"
Microsoft released a study showing the 40 jobs most at risk by AI:
February 11, 2026 at 1:50 PM
Against stiff competition, this has to be hands down the dumbest most fucking stupid claim for LLMs yet
February 11, 2026 at 11:44 AM
Reposted
LLMs good performance on medical exams does not translate to accurate performance in real-world settings (preregistered n~1,300 study). This can't be explained by current standard benchmarks for medical knowledge & simulated patient interactions.

www.nature.com/articles/s41...
Reliability of LLMs as medical assistants for the general public: a randomized preregistered study - Nature Medicine
In a randomized controlled study involving 1,298 participants from a general sample, performance of humans when assisted by a large language model (LLM) was sensibly inferior to that of the LLM alone ...
www.nature.com
February 11, 2026 at 10:30 AM
Non-consensual porn being the other major use case of course
February 11, 2026 at 11:39 AM
Nah, it's growth of leveraged ETFs, single stock futures, 0dte options
Vaguely formed hypothesis: the crypto ‘correction’ and the maturing of agentic AI are connected. At least part of the crypto thing was under-engaged tech people looking for something to do with their time. Now they’ve got Moltbook to play with, crypto has gone back in the toy box.
February 11, 2026 at 11:36 AM
Bear case for new grads is that there is so much AI slop poisoning most interactions that the risk/reward to interacting with a stranger is increasingly skewed to the downside
A bit concerned for the future of graduate roles atm. Is there a bull case?

I was lucky enough to be one of the 1200 people in 2011 hired at big four. Unlikely to be anything like that now.
February 11, 2026 at 8:55 AM
Plus slightly reduced social pressure to have kids, especially for women
Martin Wolf absolutely on the money here. Declining fertility rates are about structural changes in society- more equal opportunities.

Reform and voices on the right want to turn back the clock on women's opportunity. Progressives should give it short shrift.

giftarticle.ft.com/giftarticle/...
Why are fertility rates collapsing? Gender roles
A big part of female graduates’ decision to have children depends on how they expect their husbands to behave
giftarticle.ft.com
February 11, 2026 at 8:40 AM
Guessing "after decades and 100s of billions of dollars of investment, cars that are orders of magnitude more expensive than normal cars, driving under optimal conditions in a tiny geofenced area still need human backup" is too long to fit in a headline
February 11, 2026 at 1:37 AM
Friend of mine set a complex analysis multichoice exam where all of the 20 answers were C. Everyone failed
Ever notice that multiple‑choice quizzes often feel strangely predictable? I recently dug into why answers so often sit in the middle—and what this reveals about our deep, systematic difficulty with randomness. It’s more surprising than you might think. 🧵👇
1/10
February 10, 2026 at 8:15 AM
They seem fine with the random number generator approach to US tariffs though
if you somehow missed our weekend splash... hardly full-throated support for Starmer, but business figures and investors fret over the alternative
www.ft.com/content/c855...
February 9, 2026 at 12:05 PM
How is this news to anyone?
February 9, 2026 at 9:02 AM
This wildly overestimates the Elon Musk investor
The most likely reason Elon Musk finally disabled Starlink for Russian soldiers is because he wants to take SpaceX public and get a trillion dollar valuation for this IPO.

If the Russians were using Starlink to commit war crimes, then investors might have boycotted his IPO.
February 8, 2026 at 6:30 AM
Not exactly "reasoning", it found a match for the solution to a particular calculation in a different field. The interesting thing for me is why the Llms aren't doing this more often
My latest story for @wired.com... An AI math startup solved 4 math mysteries, a sign of the technology’s steadily advancing reasoning capabilities.https://www.wired.com/story/a-new-ai-math-ai-startup-just-cracked-4-previously-unsolved-problems/
A New AI Math Startup Just Cracked 4 Previously Unsolved Problems
Axiom says its AI found solutions to several long-standing math problems, a sign of the technology’s steadily advancing reasoning capabilities.
www.wired.com
February 5, 2026 at 1:56 PM
For those who don't know what Dubai cookies are, they look and smell like someone vomitted into a marshmellow
sure the us is sliding into authoritarianism but on the other hand, south korea is completely paralyzed by dubai chocolate fever. in a couple of weeks they went from dubai chocolate chewy cookie to dubai chocolate raw amberjack fish to dubai chocolate grilled short ribs
February 5, 2026 at 7:48 AM
Technically it's only your inmune system that it's claiming it may help reinforce
America: where you can sell a random “wellness” drink at CVS, claim it “may” somehow “reinforce” your immune system, and literally call it “The Flu Shot.”
February 5, 2026 at 5:38 AM
I remember L&G head dealer saying electronic trading of equities would never take off as he would always want the colour
February 4, 2026 at 10:39 AM
"analysts"
February 3, 2026 at 11:35 AM
Main thing Meeker taught is us that as long as you make the right people rich, it doesn't matter how many customers you impoverish
"While many of Meeker’s boldest projections seemed wildly optimistic at the time, the vast majority proved not only directionally correct but often modest in hindsight." www.bloomberg.com/news/newslet...
February 2, 2026 at 6:58 AM