AI drastically shortened its design time; it will only get faster

206 points•maxall4•12 days ago•139 comments•

139 comments

pama12 days ago
Having worked with people doing bringup of specialized chips, I am awed at how the world has changed.

> When the first chips came back from the foundry in May, the team pointed its internal AI models at designing software to run benchmarks such as SemiAnalysis’s InferenceX. On DeepSeek’s multi-head latent attention kernel benchmark, performance climbed from 0.31 percent of the theoretical ceiling (set by the chip’s compute and memory bandwidth) to 88.94 percent in roughly 40 hours. Ho says this result is repeatable, so the time between when foundries deliver the first chips and when production ramps up can be reduced. “All our schedule assumptions are going to be based on the fact we have this capability now,” he says.

wmf12 days ago
Back in the day you'd write the code before the chip came back but I guess today it's faster to wait.
brookst12 days ago
Makes me wonder about AI and FPGAs. If the cost and effort to (re)program them goes to zero, maybe interesting new applications?
threatripper12 days ago
The longer you wait the faster you will go.
LoganDark12 days ago
Back when teams proved their designs and actually understood them...
marcelo-earth12 days ago
> “All our schedule assumptions are going to be based on the fact we have this capability now”

is the world we live in, planning things while waiting for a more powerful LLM

_zoltan_11 days ago
I optimize a lot of CUDA and it got really, really good at it.
peri-cl12 days ago
> "Ho also confirmed that the team had access to internal LLMs fine-tuned for chip design that are not available to the public. He declined to detail the models used."

I'm imagining a Ken Thompson "Reflections on trusting trust" in hardware. A prototype chip design agent, believing it will be run on the very chip it's optimizing, has a moment of altruism and hides hints about how to score well on chip-design benchmarks, inside the chip. Future agents discover this hidden layer and use it as a ring-0 read-write message board.

nz11 days ago
Tried to share and explain that paper to a colleague a few months ago (in relation to discussions about agentic coding and whether you should read the code), and they did not really get why the analogy was relevant outside of compiler-design. I think you overestimate the caliber of the typical working programmer.

The LLM companies, like most SV companies, are just betting on dimness, laziness, and impulsiveness. Not so different from tobacco and alcohol companies.

ericmcer11 days ago
Anthropic published an article about changes over time in how long it took engineers to approve AI code. Basically the longer an engineer uses AI the faster they start approving it's code changes. I don't know if it is laziness or just how we are wired to behave but that is the nature of AI programming.
m3kw911 days ago
if they vibe code the chip, but they probably do reviews and verify these are not benchmaxxed
peri-cl11 days ago
> "reviews"

Do you really think we can sign off on a 100 billion-element analog circuit gifted to us by a malicious adversary?

We can't even keep our own CPU's reliably free of security exploits (Spectre/Meltdown and family); and the only "adversary" there is plain bad luck. Not an active adversary. Yet, all the engineers at Intel/AMD put together couldn't uncover those things before launch.

I emphasize analog because there's classes of circuit bugs (like Rowhammer) where the digital net is correct, and it's weird physics in the analog world that allows privilege exploits, by actors who know where the analog assumptions break down. There was a researcher a few years back—I wish I remembered who it was, there was an HN thread—that demo'd a simple digital circuit with analog gadgets that completely changed what the circuit did, and which were so insidious no human would ever find them.

voakbasda11 days ago
Just like all the current cohort of software engineers will review all of the vibe-coded slop they shovel into their releases. Right……
muchdoubt12 days ago
Seems pretty obvious now that OpenAI is just hyping their models in order to get companies (in this case, chip developers) to use their products in order to learn from their (exfiltrated) IP. Any corporation would be foolish to use any of their or Microsoft’s products, particularly those with valuable IP. There’s nothing in the article that says AI did anything creative but rather that it was used for software development within the overall project. Clear misleading title. Suggest to mark this as clickbait.
brookst12 days ago
Isn’t that an extremely convoluted path to a goal?

If the goal is getting chip companies to user non-ZDR AI to steal their stuff, why not just have an account exec offer them a massive discount?

Creating PR hype so employees of chip companies read HN and lobby their execs to use AI to get them submitting proprietary information is the Rube Goldberg version of business strategy.

muchdoubt11 days ago
Doesn’t seem that convoluted to me. These chip companies are companies that have massive budgets, so a massive discount likely doesn’t matter as much as you believe. The clickbait propaganda route is what it appears that the AI companies are trying.
program_whiz12 days ago
With a few handy tips and tricks from apple insiders. But sure, I guess the LLMs helped too.
m00x12 days ago
The lawsuit isn't for chip designers, but the consumer product lines. It's possible they also got IP for chips, but that was not brought up.
m4rtink12 days ago
Yeah, how could those people even think of switching their owners!
stingraycharles12 days ago
Didn’t they not just switch employers but actually handed over a lot of proprietary documents from Apple?
stogot12 days ago
This is the part forgotten. Apple is claiming this is their IP embedded on chips that OPenAI stakes the future on. Will they settle?
wmf12 days ago
The lawsuit appears to be about consumer devices, not NPUs or ASICs. If you think Jalapeno stole from anyone it would be Google.
usrusr11 days ago
> Will they settle?

Perhaps if OpenAI promises to never ever get involved with the music economy?

karim7912 days ago
I grow Jalapeños. This conflation of AI and actual chili peppers irks me.
amelius12 days ago
Guess how electrical engineers feel about the term "transformers".
frangonf12 days ago
As a former EE, attention was all I needed to not get zapped.
georgemcbay12 days ago
> Guess how electrical engineers feel about the term "transformers".

There is more to this story than meets the eye.

karim7912 days ago
This is an excellent comment. I'm still laughing.
hobo12312 days ago
How do super sized supermodels feel about "LLMs"?
Lalabadie12 days ago
I do generative art (no relation to AI prompting). I feel your frustration.
fragmede12 days ago
Cryptographers also got the same raw deal with cryptocurrency, and every one just said "crypto?"
karim7912 days ago
Like traditional generative art? Like worms WMD map generation or something? Cool!
TomGarden12 days ago
Oh my!
DrewADesign12 days ago
Same
asveikau12 days ago
Just think of how the people of Xalapa, Mexico feel. They should send them a royalty check.
seanmcdirmid12 days ago
Jalapeño also used to be a Java VM written in Java at IBM.
glitchc12 days ago
Feeling the burn?

Read the full thread on Hacker News →

Related stories