Richard Penner (@richardpenner) header
Richard Penner (@richardpenner) profile picture

@Figure1 co-founder

Toronto
Joined February 2009
Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)Photo by Richard Penner (@richardpenner)
The Verge article blaming browsers for a shitty mobile web is 6MB and has more than 1,000 javascript errors.
19
620
643
I forced myself to listen to this, and roughly: AI gets so smart it figures out a way around data center power cutoff. Spreads itself to laptops (but then why was the data center needed?), finds humans willing to help, yada yada yada, military overwhelmed, government overthrown. Many of the people that understand AI the deepest understand society the least. They have always better understood machines than human behavior. Nothing wrong with that, until they start fear mongering.
Dwarkesh, you are SO GOOD at what you're doing. Why don't you use that talent to paint a well-rounded picture? Why not have a single episode on this topic with someone not from Ant/OAI?
1
49
Richard Penner retweeted
If that Spy’s goal was to destabilize one of the most successful American companies, they achieved their goal. Ousting Jensen feels rushed and not right.
489
769
82
14,132
366,592
Maybe this is a scorching hot take on Twitter but frontier models with lots of code are still putting up many junior-level mistakes on the first turn (N+1 queries, duplicating logic all over the code base) and I have to assume that people saying they don’t read code anymore are OK with this. It’s sloppy technical debt, yet it may be that for many products that debt doesn’t matter.
4
5
305
Local councillors and mayoral candidates at my door handing me flyers with large fonts are leaving with a glance at @EricDLombardi’s website and a request to provide the same level of rigour to their own campaign. Sadly I think most people vote based on vibes.
1
51
He has a 35MB PDF, and strays outside of his party’s norms. More of that, please.
14
Richard Penner retweeted
What AI has done so far: MEDICINE • Helped paralysed people speak again. (Brain implants turn intended speech into words, even recreating their own voice.) • Helped a paralysed man stand and walk again. (A brain–spine interface let him control his legs by thinking.) • Helped blind people read and understand the world around them. (Describing surroundings, reading labels and identifying objects through a phone camera.) • Uncovered cancers doctors would otherwise have missed. (29% higher breast cancer detection in a major study.) • Identified antibiotic candidates that kill drug-resistant bacteria. (Abaucin targeted a dangerous superbug in lab and animal tests.) LEARNING • Made personalised learning available on demand. (An ‘Einstein’ as your personal tutor whenever you want to learn.) • Helped students learn twice as much in less time. (A custom AI tutor outperformed an active-learning Harvard physics class.) • Helped people write, code, design and build without years of specialist training. SCIENTIFIC DISCOVERY • Predicted the structures of over 200 million proteins. (Opening new paths to understanding disease and developing medicines.) • Discovered planets hidden in telescope data. (Including Kepler-90i, the eighth planet in a distant solar system.) • Recovered ancient writing buried by Vesuvius nearly 2,000 years ago. (Reading inside carbonised scrolls too fragile to unroll.) • Begun unlocking how animals communicate. (Patterns in whale calls. Evidence that elephants use individual, name-like calls.) • Produced a proposed solution to one of mathematics’ hardest problems. (Navier–Stokes: 10,000 AI agents, 88 hours, according to OpenAI.) ENVIRONMENT • Predicted where a hurricane would strike nine days before landfall. (GraphCast forecast Hurricane Lee’s Nova Scotia landfall about three days ahead of conventional forecasts.) • Engineered enzymes that break down plastic in hours rather than centuries. • Detected wildfires before the first emergency call. (Spotting smoke and alerting firefighters earlier.) • Slashed the weedkiller farmers need by targeting weeds individually. (35% less herbicide in sugarcane trials, with nearly the same weed control.) ENERGY • Advanced the science of clean fusion energy. (Controlling and shaping superheated plasma inside an experimental fusion machine.) • Driven a generational wave of investment in carbon-free energy, from nuclear power to solar and wind. (Microsoft’s Brookfield deal alone targets over 10.5 GW of new renewable capacity by 2030.) TRANSPORT • Delivered dramatically safer driverless journeys. (Waymo: 81% fewer injury crashes per mile than human drivers in the areas studied.) • Given people who cannot drive a new way to travel independently. (Including blind passengers and older people who cannot drive.) JOBS AND GROWTH • Fuelled an investment boom. (AI-related investment categories accounted for an estimated 37% of U.S. growth in the first nine months of 2025.) • Driven demand for the skilled trades building AI infrastructure. (Electricians, welders, construction crews and cooling specialists.) WHAT AI HAS NOT DONE • Replaced all the radiologists. Guess what? We still need more of them. • Used a community’s water. Golf courses use more. (U.S. golf irrigation: roughly 550bn gallons in 2020. Data centres’ direct consumption: 17bn in 2023. Microsoft’s next-generation designs consume zero water for cooling.) • Raised everyone’s household electricity bills. (Oregon’s PGE: residential bills cut 1.3% after shifting costs to data centres. Indiana’s I&M: proposed household savings of roughly $100 a year, supported by large-customer growth.) • Demonstrated that it wants to kill us all.
102
415
53
1,350
136,887
I built a proxy model extension for pi that rejects responses until they are ASD STE-100 compliant, using a fast critic model. If you have caught on that frontier models are getting worse at human communication, this is the fix.
This didn't really work. A passive instruction never does that much to alter output tone. So I'm shipping a new skill in v1.2: /wait-what For when you have no clue what the agent is saying
1
194
Richard Penner retweeted
Adding to every AGENTS md file for the rest of time. (h/t @richardpenner for the self-referential explanation on what ASD-STE100 is)
The fix for this is to say: only report to me in ASD-STE100 Simplified Technical English
54
106
18
3,835
520,919
Richard Penner retweeted
Replying to @JensenHuang @nvidia
880
2,749
1,106
30,041
10,530,276
Richard Penner retweeted
Lots of very smart people are appropriately concerned about regulatory capture from top two AI players. And there are many scary press releases but no real due process. If OpenAI violated Hugging Face in a way that “demands” new regulation; let’s start with a formal criminal investigation. With a proper third party investigation. And sincere liability. It can’t be “so super serious” and only analyzed by the defendant. Thats ludicrous. And why would we write new regulation if the crime we are supposedly protecting against isn’t even prosecuted? This stuff makes no sense.
118
239
35
2,338
170,698
Richard Penner retweeted
Replying to @deanwball
The "ungovernability" (and openness) of Linux and the Internet is precisely what has made their success. The same will be true of open weight AI foundation models.
79
397
51
3,312
183,222
Richard Penner retweeted
open-sourcing a 3D gen toolkit for Claude Code input image → environment, meshes, physics, lighting, & audio
108
387
79
3,547
408,684
Richard Penner retweeted
Marcus Hutchins, the guy famous for stopping the WannaCry Ransomware, probably has the best take on Mythos doing vulnerability research
101
533
88
4,219
513,768
when you are vibe coding and you finally try the app after an hour
326
2,365
242
33,974
2,755,956
Project Hail Mary deserves the 70mm IMAX experience. Incredible movie.
1
774
This is a terrible idea for a variety of reasons. Governments need to focus on the basics. Expand electricity supply at lower prices. Fast or by-right permitting for major industrial and data center infrastructure. Get the fundamentals right and Canada will have plenty of domestic competition for AI services and autonomous machine manufacturing. A recommendation for nationalization to have services *several years behind* leading models is not going to cut it.
11
11
2
150
13,610
Weekend project: took just a few hours to deploy an Arduino, Raspberry pi, CT clamp, a few resistors and capacitors, and a custom built dashboard for monitoring my sump pump. Codex was even deploying the Arduino software via the pi.
1
3
650
Richard Penner retweeted
The way society has allowed geopolitics abroad to turn us into animals at home is abhorrent. Jewish Canadians are our friends and neighbours, and we should all stand against what's happening to their community here. People are more than their ethnicity. Today you, tomorrow me.
51
46
6
527
28,466
One of my favourite Chat GPT hacks: a custom GPT for every product I own. When I get new appliances, electronics, even vehicles, I find the instruction manual PDF and create a custom GPT. This even works for complex board games, where you can simply say "Help us setup the board"
3
2
267