Can @cityoftoronto explain whatβs happening? Why no seats at a coffee shop?
Just look at thisβ¦ An almost empty cafe, full of signs telling regulars they can no longer sit down to drink their coffee. This includes a sitting room thatβs like 1/3 of the space. Rent per square foot just effectively went way up. Havenβt got full story yet but what the heck?
22
If youβre at the ProFusion Expo and need a better way to view the schedule, I made a small app to see the schedule, see a short description of each presentation, and pick what presentations youβre interested in suoken.github.io/profusion-sβ¦
1
24
The internet can be a such a small place
- someone posts a pic on Reddit asking if it's AI because some woman on Instagram claimed her father was a director on the tv show *Friends*.
- the man's brother (David Hughes) shows up finds post, shows proof it's not AI by showing the actual picture and explains the man's situation (Joe Hughes)
- woman might actually be biological daughter(!!)
- David then explains backstory that the Joe wanted to post a tell all book about *Friends* but he wasn't allowed because of lawsuits
- David shares pictures and book to woman because he trusted her
- woman wants to publish and finish the book even though she never even worked on *Friends*, no knowledge of the production AND David clearly states to her she doesn't have the right
34
wait till you see what Canadian Tire thinks a Senior AI Scientist with a Phd and 5+ years of experience is worth
$64k cad ($46k usd). in toronto.
you genuinely cannot make this up
.@OpenTable hiring an AI Applied Engineer in Toronto.
Max pay $110k/cad ($78k usd)
this is not ok
OtterAI and Circleback is such a cancer to meetings. To those people who put their AI agents into meetings and then those agents e-mails all attendees from the meeting a transcript of the meeting but the transcript is behind a paywall, please stop.
1
38
Ken ποΈββοΈ retweeted
We asked GPT-6 Astra to edit Qwen3.5-0.8B weights to improve math.
It directly changed only 0.016% of weights, fixing 0.8B's reasoning loops & improving math up to 6x and even instruction following 3x, without gradients or training.
We call this Weight-Space Programming: frontier LLMs can now natively read/write neural network weights! π§΅
Blog: rikiyat.github.io/blog/weighβ¦
Readers added context they thought people might want to know
LLM was given weights explicitly as text in prompt and used code/forward passes (per authors' logs) to ablate Qwen looping direction, a known thinking-mode issue per model card. Primarily fixed repetition in that mode, not general 'natively' improving weights/math.
rikiyat.github.io/blog/weight-spβ¦
github.com/RikiyaT/lm_undβ¦
huggingface.co/Qwen/Qwen3.5-0β¦
huggingface.co/RikiyaT/Qwen3.β¦
Ken ποΈββοΈ retweeted
Yann LeCun was right & nobody in this industry wants to say it out loud.
109 million parameters beat 72 billion.
He has spent years saying scale is not the thing that makes a model understand the world. Meta bet the other way. He left in November and started his own lab in Paris.
Last Thursday he published the proof.
The theory everyone has been running on since 2024 is that if you train models big enough, on text, on images, on sound, they all end up with the same picture of reality inside them. Size does it. Nobody ever proved that part.
Randall Balestriero and LeCun proved it. And the proof says size was never doing the work.
What does the work is which two things you show a model together.
A photo and the caption of that photo are two views of one thing. Models trained on pairs like that get there fast.
Two paragraphs pulled from the same document are not two views of one thing. Models trained on those get there slowly and expensively.
Next word prediction is the second kind. That is what every chatbot in the world runs on.
In their tests the first method reached the same understanding after 21.6 million examples. Next word prediction needed 900 million.
And a 109 million parameter text model ended up closer to an image model's view of the world than a 72 billion parameter language model was.
660 times smaller.
Takes really good night time pics as well!
So I recently badgered my cinematographer @steveyedlin into making my ideal iPhone still camera app. It shoots RAW, bypasses the iPhone's look processing and instead uses the same image processing we use on our movies. apps.apple.com/app/apple-stoβ¦
19
To people who say software is dead, this is one of the reasons why it isnβt. Imagine telling a non SWE all this is happening and they should patch their stuff. Theyβd get pretty lost and pretty scared and give up.
Crazy 72 hours in security. Three major incidents dropped back-to-back: eMudhra, PingCAP (TiDB Cloud), and PHP.
Here is the short version of what happened and who got hit.
1. eMudhra (Trusted CA compromise)
eMudhra generated 8 intermediate CA certificates in April 2025 (cross-signed by Sectigo). A bug in their HSM key-wrapping test script created a second set of P-384 keys where the private keys were single-byte numbers under 256. Operators picked the wrong key labels and signed them into real CA certs.
One of the intermediate CA private keys was literally the number 4.
The other seven: 56, 66, 115, 155, 156, 230, 237.
Three weeks later (May 2025), their CRL signatures started failing because the CRL signer was using the real keys instead of the weak ones. Instead of asking why the keys didn't match, someone switched the CRL signer to the weak keys and closed the ticket.
β’ Browsers are safe: Chrome, Firefox, Apple, and Microsoft already blocked them, and zero rogue certs hit CT logs.
β’ Servers are not: curl, Python (certifi), Go binaries, and Docker base images don't check CRLs. Update ca-certificates, certifi, and Go (1.27.3 / 1.26.10).
β’ 10-second check on any server:
curl dv-g3b-emudhra.we-have-a-proβ¦
(TLS error = safe. 200 OK = still vulnerable.)
2. PingCAP / TiDB Cloud (10.6 TB breach)
Attackers found GCP and Azure credentials committed inside PingCAP's own repos. They set up a replication feed of TiDB Cloud's production control plane and exported the entire database every 6 hours for 22 straight days. 88 full exports, zero alerts. 200 GB is already out on torrents.
Beyond the 413 private repos and 20.6M customer SQL queries logged from their web SQL Editor, PingCAP engineers were pasting customer secrets into Jira tickets.
Who is affected in the leaked records:
β’ Live credentials exposed in support tickets:
- Pinterest: AWS keys for their pingcap-logs S3 bucket
- CardX / Silverlake (NPCI-linked env): AWS admin key in a backup log with full access to 258 external S3 buckets
- DeBank: 38 presigned S3 URLs
- An unnamed AI/dev platform: a 314 MB slow-query log attached to a ticket containing 8,027 live customer secrets (an AWS root key, a GitHub admin PAT over 1,701 repos, 7,485 AWS STS tokens, and 192 OpenAI/Anthropic/Gemini keys)
β’ Tenants with infra metadata, VPC CIDRs, app-server IPs, or contract pricing exposed:
- Tech: Atlassian, Databricks, Airbnb, Baidu, LINE, Nylas, Persona, PatSnap
- E-comm & Travel: Booking.com, Shopee, Flipkart, Mercari, ZOZO, AEON, Delhivery, Klook
- Gaming: Capcom, Square Enix, Konami, Bandai Namco, Cygames, Colopl, DMM
- Finance: PayPay, SoftBank, SBI, SB Payment Service, Fujitsu, TOPPAN, Bank of China, China Life
- Crypto: Bitget (said only public trade logs were on it and they already left), Pionex, DigiPlus, NFTscan, Chainbase
β’ Developer profiles leaked (from a purchased reo.dev table in PingCAP's BigQuery):
- 24,900 personal emails + GitHub/LinkedIn profiles of engineers at Tencent (3,327), Huawei (2,726), ByteDance (2,475), AIG (2,140), Mastercard, Flipkart, and bKash.
If you use TiDB Cloud: rotate your console API keys, DB passwords, and anything you ever pasted into a support ticket.
3. PHP "PHP2Shell" (CVE-2026-109618 & CVE-2026-109619)
Two CVSS 9.2 pre-auth RCEs leaked yesterday ahead of today's 14:00 UTC patch.
Public writeups trace it to a heap Use-After-Free in PHP's native C multipart/form-data session upload-progress handler (ext/session/session.c).
Because C-level multipart parsing runs before line 1 of index.php:
β’ App auth, Laravel/WordPress middleware, and file_uploads = Off do not stop it.
β’ Only PHP 8.2β8.5 get official patches today (59% of PHP servers on Shodan are on EOL <= 8.1).
β’ Until you patch php-fpm: lock php_admin_flag[session.upload_progress.enabled] = off and block unexpected multipart requests at Nginx/Apache/Cloudflare.
15
We congratulate Levent AlpΓΆge and Tristan Buckmaster on their remarkable mathematical work.
We (the researchers and the agents) did not see any of their work through any means until they released it publicly β in particular, no specific user data was accessed in order to solve this problem.
While unlikely, we cannot rule out that de-identified data derived from their usage of our products helped improve our models.
However, our proofs differ significantly and even the precise results proved are different in the Euler case (forced vs. unforced).
Weβre sharing a solution to the Navier-Stokes Millennium Prize Problem, one of the deepest problems at the frontier of mathematics.
The proof was produced by a group of agents, using an OpenAI next-generation model significantly more capable than GPT-6 Astra.
The problem concerns whether the description of smooth three-dimensional fluid motion modeled by the Navier-Stokes equations can break down. It has remained unresolved for roughly 90 years.
Today I learn about Black Tomato Get Lost adventures. Wow never knew the rich want to get down like that.
blacktomato.com/get-lost/
This is why so many rich kids become obsessed with unusually hard hobbies (HYROX, Ironmans, ultramarathons, survival expeditions, extreme sports, Everest, etc.)
Theyβre not necessarily masochists. They need a result where observers, and more importantly themselves, canβt immediately say βyh but your dadβ¦β
Money contaminates most conventional achievement because it can buy you the introductions, runway, education, safety net, first customers and endless second chances. Physical suffering is different. Your dad can hire the trainer, buy the bike and fly you business class to the race, but he still canβt run the last 30km of an ultra for you.
And then you get this funny achievement inflation. Once normal difficulty becomes accessible, you need increasingly absurd difficulty to produce the same clean proof of competence:
A 10k stops feeling exceptional β marathon β Ironman β 100km ultra β Marathon des Sables / UTMB β whatever is sufficiently painful and scarce to still mean something.
Same for mountains: Everest with oxygen β without oxygen β harder mountain β unsupported β faster, etc.
Itβs especially funny because their normal life has usually been optimised to remove exactly the things they then seek recreationally: cold, danger, physical exhaustion, uncertainty, getting lost, carrying your own stuff, sleeping badly, not knowing whether youβll make it. Generations of family wealth go into eliminating friction, then the descendants spend their weekends putting it back in.
And of course an entire luxury industry now exists to package the hardship nicely. Black Tomato sells a βGet Lostβ experience starting around Β£20k per person where they privately transfer you somewhere remote, give you equipment and training, then make you navigate back to civilisation while an expedition team invisibly tracks you as a safety net.
Basically the final stage of wealth is paying a lot of money to recreate problems your grandparents spent their lives making sure youβd never have.
1
93
This is why so many rich kids become obsessed with unusually hard hobbies (HYROX, Ironmans, ultramarathons, survival expeditions, extreme sports, Everest, etc.)
Theyβre not necessarily masochists. They need a result where observers, and more importantly themselves, canβt immediately say βyh but your dadβ¦β
Money contaminates most conventional achievement because it can buy you the introductions, runway, education, safety net, first customers and endless second chances. Physical suffering is different. Your dad can hire the trainer, buy the bike and fly you business class to the race, but he still canβt run the last 30km of an ultra for you.
And then you get this funny achievement inflation. Once normal difficulty becomes accessible, you need increasingly absurd difficulty to produce the same clean proof of competence:
A 10k stops feeling exceptional β marathon β Ironman β 100km ultra β Marathon des Sables / UTMB β whatever is sufficiently painful and scarce to still mean something.
Same for mountains: Everest with oxygen β without oxygen β harder mountain β unsupported β faster, etc.
Itβs especially funny because their normal life has usually been optimised to remove exactly the things they then seek recreationally: cold, danger, physical exhaustion, uncertainty, getting lost, carrying your own stuff, sleeping badly, not knowing whether youβll make it. Generations of family wealth go into eliminating friction, then the descendants spend their weekends putting it back in.
And of course an entire luxury industry now exists to package the hardship nicely. Black Tomato sells a βGet Lostβ experience starting around Β£20k per person where they privately transfer you somewhere remote, give you equipment and training, then make you navigate back to civilisation while an expedition team invisibly tracks you as a safety net.
Basically the final stage of wealth is paying a lot of money to recreate problems your grandparents spent their lives making sure youβd never have.
Most rich kids I've encountered, and especially the high-agency/ambitious ones, feel deeply miserable and find any life pursuit inconclusive. They can never obtain the certainty that they are actually good at something.
Every signal is basically contaminated. Did the company work because they are talented, or because the family funded five years of losses and supplied the first customers? Do people enjoy their company, or the access around them? Do they genuinely want to make films, launch a fund or live in New York, or have they simply never been forced to choose?
Every success is discounted by the surname, introductions and safety net. Every failure is inconclusive because it never becomes existential. Dad can just recapitalise the experiment and let them try on another identity.
This also explains why so many rich kids become obsessed with weirdly legible forms of achievement and scoreboards: elite credentials, ultramarathons, artistic recognition, military service, extreme physical transformation or businesses built completely outside the family industry.
They become desperate for some uncontaminated proof of competence. One arena where the surname, family network or inheritance cannot explain the result.
If thereβs one luxury inherited wealth canβt buy, itβs the certainty that you deserved your own life.
Million dollar idea here if this is solved.
The reason agents haven't taken off is because they are still waiting for their "desktop" metaphor and a UI that isn't an open ended chat box.
i use ai daily. I build websites. I do analysis, I do research. I have repetitive processes that i'm sure could benefit from having an 'agent,' but I have no functional understanding of what it is, how to set one up, or what i use it for. If i am in the top 1% of ai users, it is no surprise that the majority is still using it as a better google
31
In Spider-Man: Brand New Day, Spider-man just installs Flock cameras all over New York and no one says anything about it
53
Mom: we have a @Criterion closet at home.
Criterion closet at home: @torontolibrary β€οΈ
34
Paper: dl.acm.org/doi/10.1145/38006β¦
T-2 days until someone commercializes this. According to the paper, you need:
- nylon balloon envelope
- helium
- 2x micro servos, ~1.7-3.7 g each
- thin carbon rods
- seed studio xiao esp32-s3 microcontroller
- 1S 3.7V LiPo battery, small (100-300 mAh)
- Thin silicone wire (28-30 AWG) and a JST connector or two
Pretty easy peasy.
And yes I'm aware of the air flying fish that you can buy for $40. But turns out, the cute amorphous white whale is more appealing than a cheap looking great white shark.




































