AI 行业热点

🎙️ 播客精选

Why Hardware-Software Co-Design Is AI’s Real 100x: Dylan Patel of SemiAnalysis

Training Data · 2026-06-30

Speaker 1 | 00:00 - 00:24
I think it’s really fun inside of Semi Analysis because we have 90 people, and a big chunk of them are technologist engineers across the whole supply chain, and then a big chunk is people who are formerly at hedge funds. And you see these arguments, people are like, Oh, that doesn’t matter. Someone’s like, well, but cost? And then the engineer’s like, no, no, no, but this technology is the coolest. And you see this organically fight it out, and we’re pretty informal.

Speaker 1 | 00:24 - 00:27
And given the fact that I was a former moderator, you can imagine what

Spe…

🎧 收听完整节目

🐦 X/Twitter 热点

Swyx (@swyx)

  • very successful first poster sessions day at AIEWF, thanks to @heathercmiller and @ACM_President for the support!!

tomorrow: Poaster sessions! submit your hottest tweets to @vibhuuuus for printing!

spotted: left shark [40 ❤️ 4 🔄]

Boris Cherny (@bcherny)

  • Agree [797 ❤️ 9 🔄]
  • You asked, we listened. Claude Desktop on Linux is here!

Download link: [3503 ❤️ 196 🔄]

Peter Yang (@petergyang)

  • What does this mean btw? After I hit 50% weekly usage Fable is no longer available? [379 ❤️ 6 🔄]
  • btw I have no idea if it’s coming out midnight or not [6 ❤️]
  • Setting an alarm at midnight to get Fable to fix all my projects 😅 [136 ❤️]

Nan Yu (@thenanyu)

  • tru tru [4 ❤️]
  • The definition of “distillation” is interesting. Going by this logic, the entirety of Cursor’s training data was distilled from Claude in the early days [47 ❤️]

Madhu Guru (@realmadhuguru)

  • The biggest challenge for trad PMs trying to adapt to AI native building is a lack of magical thinking. A decade of frameworks, “agile” and metric obsession has led to constraint-first, incremental thinking.

I used to combat this on my teams by asking them to imagine we had tech from 100 yrs in the future, the experiences it would enable and working backward from that. the ideas would always be far more creative.

the funny thing is, that technology from the future is here.

go build that magical thing! [56 ❤️ 4 🔄]

Thariq (@trq212)

  • making some last minute changes to my deck- see you tomorrow at AIE! [467 ❤️ 12 🔄]
  • And as we say in our blog, we’re continuing to refine these safeguards to better distinguish genuine misuse from legitimate requests and reduce false positives. [203 ❤️ 2 🔄]
  • Have seen some questions about the updated classifiers and wanted to clarify.

As with the original classifiers, a small fraction of routine coding and debugging tasks will be flagged and fall back to Opus.

We’re excited for guys to get access back tomorrow. [1350 ❤️ 88 🔄]

Amjad Masad (@amasad)

  • World’s first subway hackathon! [156 ❤️ 3 🔄]
  • AI is expensive to run partly because most workloads today run on generic hardware designed pre-LLMs. Etched is the first system designed from the ground up for modern inference. [1419 ❤️ 75 🔄]

Guillermo Rauch (@rauchg)

  • At dinner, tech executive is relaying his company’s @vercel feedback, and then his 12-year-old son’s @vercel feedback 😁

Vercel is for everyone. [540 ❤️ 2 🔄]

  • So great to get to work with @tobi and the wonderful @shopify team.

Excited to push the agentic web forward. [264 ❤️ 9 🔄]

  • Vercel Services

You can now collocate e.g.: a Python backend API, an ExpressJS server, and a React SPA in one Vercel project.

tl:dr;
▪️ You can run all locally with 𝚟𝚌 𝚍𝚎𝚟
▪️ Deploy and rollback all at once
▪️ Observe, monitor, debug together
▪️ Internal networking [1046 ❤️ 31 🔄]

Aaron Levie (@levie)

  • Things seem to be ending up in a better spot with Fable, and presumably GPT-5.6 next. What we have now is the initial precedent for what frontier model releases (or at least those that have significant coding and cyber capabilities) could look like going forward. This would presumably apply to bio and other categories of risk that have been identified by AI safety groups.

From the Anthropic post:

“3. A shared industry framework. Although we have reached a constructive resolution, these events have made clear that the industry needs a consistent way to assess and fix potential “jailbreaks” of AI models (techniques that bypass a model’s safeguards).2 A shared standard for judging the severity of a given jailbreak would help AI developers triage new findings as they arise, launch highly-capable models with greater safety, and communicate the level of risk consistently to government and industry partners. Together with Amazon, Microsoft, Google, and other Glasswing partners, we’ve started to develop such a framework, and we outline it below.

  1. Deeper government collaboration. We’re also strengthening our level of collaboration with the US government on new pre-release testing, information sharing, and research collaboration. We describe this deeper collaboration in the final section.”

It’s been a messy process to get here, but at least there’s some semblance of a framework that could be practical. The only note of caution here would be that there’s a lot of subjectivity that goes into various risks and their actual levels of exploitability in practice. We’re likely going to be living with a framework that requires heavy judgment and back and forth between labs and the government for major releases.

The best we can hope for is that this is a relatively efficient process, and hopefully has ways of being sped up for incremental version updates in models. It would be a bad outcome if every release after this level of threshold of capability required the same review process, and we don’t get the same rate of breakthroughs we’ve been seeing. [58 ❤️ 3 🔄]

  • We’ve been running Anthropic’s Claude Sonnet 5 through the Box AI Complex Work Eval, our agentic benchmark that puts models through real enterprise document work end-to-end.

Sonnet 5 holds frontier-class quality on complex multi-step work and pulls ahead of Sonnet 4.6 in several core enterprise domains like Energy (+4.7pp), Retail (+4.4pp), and Professional Services (+2.6pp), and other spaces where unstructured data is heavily complex.

Here are a few examples of wins compared to Sonnet 4.6 to get a sense of some of the more advanced reasoning capabilities in Sonnet 5:

  • Financing due diligence: It computed the company’s liquidity and leverage ratios from the raw balance sheet, and caught that the source report’s own stated debt-to-equity figure understated the leverage, flagging all three loan covenants as violated, not just the ones the document admitted.

  • Overhaul cost analysis: It scoped “total cost” to the company’s own KPI definitions, correctly separating out Lost Production Cost because the guidance said to track it separately rather than naively summing every number on the sheet. It also caught and handled a broken reference cell in the spreadsheet.

  • SKU revenue analysis: On segmented sales data, it computed each product’s contribution against the correct subcategory denominator, sidestepping the easy mistake of dividing by the category total, and flagged why no Pet-category SKU cracked the top 9.

Sonnet 5 will be available in the Box AI Studio shortly for customers to build custom agents with. [137 ❤️ 6 🔄]

  • More data is showing the opposite of what many people expected with AI adoption and jobs. Ramp found that the more AI adoption a company has the more their headcount grows.

At Box, we recently did a survey of 1,600+ mid and large sized companies, and the findings were similar. 58% of respondents expected headcount to rise over the next three years. Interestingly, that figure climbs to 79% among the most mature adopters of AI. The more advanced AI adopters expected to grow their headcount at a greater rate in the future than others.

Of course it’s true that the companies that can afford to adopt AI the most are also the ones that likely are seeing growth in their business, leading to more headcount. So the point of the story isn’t necessarily that by adopting AI you will inherently grow.

But the most important takeaway is that the opposite is not proving out. The fears a couple years ago would have been that the companies adopting AI the most would be hiring fewer people.

But in reality this is what actually you should expect to happen. If a company can get more customers because they use AI in sales for account or market intelligence, they hire more sales people not fewer. If you can build way more software than before, you end up hiring more engineers because the projects get bigger and you take on more. And so on. [234 ❤️ 48 🔄]

Garry Tan (@garrytan)

  • We want a California that works. [185 ❤️ 7 🔄]
  • Gbrain is mostly useful at 10,000+ markdown files in your personal brain or company brain [728 ❤️ 52 🔄]

Matt Turck (@mattturck)

  • LE DICTATEUR #fraswe [85 ❤️ 3 🔄]

Zara Zhang (@zarazhangrui)

  • “Taste isn’t valuable because it’s impossible to copy. Taste is valuable exactly because it defines what everyone else chooses to copy.” [57 ❤️ 7 🔄]
  • How can someone be THIS early [117 ❤️ 6 🔄]
  • [6 ❤️]

Nikunj Kothari (@nikunj)

  • Ok a day early. This is coming tomorrow! [2 ❤️]
  • Every developer in SF right now running back to their desks post AIEWF after they heard Fable 5 is coming back.. [5 ❤️]

Peter Steinberger (@steipete)

  • 🙃 [124 ❤️ 4 🔄]
  • Price per token != cost per task [1072 ❤️ 34 🔄]
  • Apparently we didn’t talk enough about w̶o̶r̶k̶f̶l̶o̶w̶s̶ loops yet! See ya there! [212 ❤️ 16 🔄]

Dan Shipper (@danshipper)

  • LETS FUCKIN GO [173 ❤️ 2 🔄]
  • me, on a vacation in mexico, hearing that fable might be coming back tonight [39 ❤️ 1 🔄]
  • if this happens im livestreaming and ripping tokens from my vacation in cabo tonight [150 ❤️ 2 🔄]

Aditya Agarwal (@adityaag)

  • It is a very strange state of the world where the models powering innovation in the USA are Chinese open source models. [284 ❤️ 28 🔄]

Claude (@claudeai)

  • Sonnet 5 is now the default on Free and Pro, and available to Max, Team, and Enterprise users.

It’s live across all Claude apps and the Claude Platform today, with introductory pricing through August 31: [2060 ❤️ 82 🔄]

  • Early access partners found Sonnet 5 finishes complex tasks where previous Sonnets stopped short, checks its own output without being asked, and does all its agentic work at an attractive price point. [2227 ❤️ 46 🔄]
  • Sonnet 5 is a substantial improvement over Sonnet 4.6 on reasoning, tool use, coding, and knowledge work.

Its performance is close to Opus 4.8, at lower prices. [4607 ❤️ 339 🔄]

由 Follow Builders 自动生成 · 2026-07-01