Pinned Tweet
Bend 2 is here!
It is a new programming language that blocks AI mistakes via *proof checking* - the same technique big AI labs used to solve open math problems, like Navier-Stokes.
It is also very fast, and runs on GPUs.
Watch the video. Link in the comments.
It is a new programming language that blocks AI mistakes via *proof checking* - the same technique big AI labs used to solve open math problems, like Navier-Stokes.
It is also very fast, and runs on GPUs.
Watch the video. Link in the comments.
RELEASE DAY
After almost 10 years of hard work, tireless research, and a dive deep into the kernels of computer science, I finally realized a dream: running a high-level language on GPUs. And I'm giving it to the world!
Bend compiles modern programming features, including:
- re support
- Unrestricted recursion and loops
- Fast object allocations of all kinds
- Folds, ADTs, continuations and much more
To HVM2, a new runtime capable of spreading that workload across 1000's of cores, in a thread-safe, low-overhead fashion. As a result, we finally have a true high-level language that runs natively on GPUs!
Here's a quick demo:
658
1,301
11,528
2,147,393
"when lockpicks are outlawed, only outlaws will have lockpicks"
The only way out is through.
They're calling it the Bugpocalypse. LLMs are being used to find long-hidden cracks in the software infrastructure, and it's happening at a furious rate. 3% of all compromises in computing history happened in the last week, claims one source? I believe it.
There are people saying public access to the frontier models ought to be locked out, or heavily guardrailed, over this. The problem with that idea can be summed up in one sentence: when lockpicks are outlawed, only outlaws will have lockpicks.
The lockpicks will be Chinese open-weight models, if nothing else. There's no way to stop those from becoming ubiquitous.
There are two futures in front of us. One in which only the bad guys have lockpicks, and security holes go undiscovered until they're exploited. Another in which everybody has lockpicks and is aggressively red-teaming their own software so it gets fixed before the holes are exploited.
The corpus of human code is huge but finite. The number of defects in it has to be finite too. We can fix those defects now; we have a million friendly robots to do it with.
The only way out is through.
10
15
423
9,735
in 3 days I'll be presenting Bend on the Lean Meetup, SF
bring your hardest questions!
link in the comments below ↓
bring your hardest questions!
link in the comments below ↓
44
48
1,442
24,377
Taelin retweeted
Just found out about Bend, so I had Claude make an explainer on why agents might prefer it
Every frame was drawn in Bend. Claude built a custom renderer in Bend to make the video
Every frame was drawn in Bend. Claude built a custom renderer in Bend to make the video
39
81
1,416
48,010
"Why is formal verification exploding NOW?"
1. Cost of proofs is going DOWN (as AI gets more capable)
2. Cost of exploits is going UP (as AI gets more dangerous)
3. Proofs exterminate exploits
4. Bonus: proof checking = perfect math/coding RL reward
The rest is economics
1. Cost of proofs is going DOWN (as AI gets more capable)
2. Cost of exploits is going UP (as AI gets more dangerous)
3. Proofs exterminate exploits
4. Bonus: proof checking = perfect math/coding RL reward
The rest is economics
53
124
2,150
53,612
Taelin retweeted
turns out bend was ahead of the curve
New on our Frontier Red Team blog: using Claude to formally verify software at large scale.
Formal verification can prove that code is free of certain bugs, but historically it has only been feasible for small codebases (or required a lot of expert time).
With AI models, however, this is beginning to change. In collaboration with @theoremlabs, we used Claude to prove security properties in over one hundred real-world open-source libraries over the span of a few weeks.
Read our research in full here: anthropic.com/research/forma…
25
33
1,382
66,566
hi so is anyone here doing research on AutoResearch with weights training in the loop? i.e., instead of a static model like Fable or GPT optimizing your target, you actually RL the model itself on the best results of your optimization runs? how well that works? resources?
55
8
591
30,138
Taelin retweeted
At this point, if you’re in software and aren’t currently working on a plan for moving to formal verification, it’s irresponsible.
The writing is on the wall. This needs to be your priority now.
The writing is on the wall. This needs to be your priority now.
New on our Frontier Red Team blog: using Claude to formally verify software at large scale.
Formal verification can prove that code is free of certain bugs, but historically it has only been feasible for small codebases (or required a lot of expert time).
With AI models, however, this is beginning to change. In collaboration with @theoremlabs, we used Claude to prove security properties in over one hundred real-world open-source libraries over the span of a few weeks.
Read our research in full here: anthropic.com/research/forma…
52
96
2,114
198,273
Taelin retweeted
@codegirl007 its a weird ground that is neither tests nor asserts
# Check the actual meaning of inside, independently of the movement code.
# U32 coordinates are unsigned, so x >= 0 and y >= 0 are already guaranteed.
law inside_is_correct:
for width: U32
for height: U32
for x: U32
for y: U32
{Game.inside(width, height, Game.Point{x, y}) ==
((x < width : U32) && (y < height : U32)) : Bool}
Here is an example of saying that being "inside" means that you are within the game. Now a unit test would specify hard values, where as bend is saying "for all values of width, height, x, and y"
Here is a more crazy one
# From the real starting position, every finite sequence of keys stays inside.
# Any intermediate position is covered too: its input prefix is a sequence.
# Together with inside_is_correct, this says x < 20 and y < 10.
law never_crosses_walls:
for keys: List<Char>
{Game.inside(20, 10, Game.replay(keys)) == True{} : Bool}
For all possible movement sets our character "stays" within the game.
Now this is the law, not the proof (the proof is hideous) but the law makes sense. Its for "all known movement combinations" not just "what if i was on the very right edge and i went over once more"
There you go :)
# Check the actual meaning of inside, independently of the movement code.
# U32 coordinates are unsigned, so x >= 0 and y >= 0 are already guaranteed.
law inside_is_correct:
for width: U32
for height: U32
for x: U32
for y: U32
{Game.inside(width, height, Game.Point{x, y}) ==
((x < width : U32) && (y < height : U32)) : Bool}
Here is an example of saying that being "inside" means that you are within the game. Now a unit test would specify hard values, where as bend is saying "for all values of width, height, x, and y"
Here is a more crazy one
# From the real starting position, every finite sequence of keys stays inside.
# Any intermediate position is covered too: its input prefix is a sequence.
# Together with inside_is_correct, this says x < 20 and y < 10.
law never_crosses_walls:
for keys: List<Char>
{Game.inside(20, 10, Game.replay(keys)) == True{} : Bool}
For all possible movement sets our character "stays" within the game.
Now this is the law, not the proof (the proof is hideous) but the law makes sense. Its for "all known movement combinations" not just "what if i was on the very right edge and i went over once more"
There you go :)
9
5
165
13,920
recent events motivated me to post a super quick challenge:
bank.bend-lang.com
I'll send $1k to anyone who hacks this "bank" app by either:
1. leaking the secret word stored on the `taelin` account
2. or spending someone else's money
3. or creating money out of thin air
bank.bend-lang.com
I'll send $1k to anyone who hacks this "bank" app by either:
1. leaking the secret word stored on the `taelin` account
2. or spending someone else's money
3. or creating money out of thin air
75
41
1,368
129,987
hash of the secret word:
7f36997fe0245c01ed7f8603d111797afdba42d7b7c9fcf7d98dd46deb44db76
the word is in our database, on the `taelin` account. leak it!
7f36997fe0245c01ed7f8603d111797afdba42d7b7c9fcf7d98dd46deb44db76
the word is in our database, on the `taelin` account. leak it!
6
0
174
20,082
or just follow me to do the thing that everyone will be doing 1 year later!
New on our Frontier Red Team blog: using Claude to formally verify software at large scale.
Formal verification can prove that code is free of certain bugs, but historically it has only been feasible for small codebases (or required a lot of expert time).
With AI models, however, this is beginning to change. In collaboration with @theoremlabs, we used Claude to prove security properties in over one hundred real-world open-source libraries over the span of a few weeks.
Read our research in full here: anthropic.com/research/forma…
32
28
1,216
31,915
the thing I'm cooking right now should be out before February btw
it is 5x cooler than Bend 2
(you know how my deadlines work tho but this time I think it is calibrated)
it is 5x cooler than Bend 2
(you know how my deadlines work tho but this time I think it is calibrated)
5
0
144
4,791
