WSJ tech columnist and author of Arriving Today, about the insane, around-the-world journey all the stuff you ordered takes on its way to your front door.
Baltimore, MD
Joined March 2007
- Tweets 476
- Following 1,586
- Followers 97,682
- Likes 217,666
Bold Names is our li’l interview show at WSJ, send me a question at Christopher dot Mims at WSJ if you want to be on an upcoming episode
Christopher Mims 🤌 retweeted
This week on Bold Names, @mims & I talk with Uber Chief Product Officer Sachin Kansal about the robotaxis.
Christopher Mims 🤌 retweeted
The funny thing about this is that it seemed like a tee-up. Ask the question everyone is wondering about figuring Sam would have a quick stock answer ready to go.
Christopher Mims 🤌 retweeted
The kinds of sophisticated AI models that companies are using to get real work done and reduce head count aren’t the ones getting all the attention, writes Christopher @mims on.wsj.com/43FyT3T
Christopher Mims 🤌 retweeted
What if fracking but ... renewable?
NREL estimates there is enough geothermal energy beneath our feet to power the entire country thousands of times over.
Why haven't we tapped it until now? It cost too much. But now that new power plants are $$ ...
wsj.com/business/energy-oil/…
Christopher Mims 🤌 retweeted
Improbably, there is one type of renewable energy which got a huge boost under Biden, saw its tax incentives preserved under Trump, and is backed by the likes of Bill Gates. And it's finally ready for prime time.
It even uses made-in-USA tech from the oil and gas industry!
good thing no one's unaccountably killing giant energy projects because they don't like how the energy is made
A very informative interview with a $META employee working on infrastructure on what the state of the market is right now:
1. The number one constraint for $META right now is the availability of the required energy capacity. GPUs are not a constraint as they have been planned ahead already, but he does mention that the energy capacity constraint will impact future GPU orders. He does mention that this problem is industry-specific and not $META specific.
2. Another constraint not specific to Q3 is the supply of custom silicon that can be purchased from $TSM. Long-term 2nm supply is going to be a problem for $TSM to meet.
3. He expects that HBM producers like $MU, Samsung, and SK Hynix will continue to demand high premiums for at least the next 3-5 years, as HBM is one of the most premium-based commodities that is required.
4. Specifically for AI inferencing, there is a huge dependency on SSD and HDD storage.
found on @AlphaSenseInc
Christopher Mims 🤌 retweeted
Thus queuing up one of my favorite times I have been quoted.
...and here's the full interview
piped.video/watch?v=tpeV-ZrL…
Smartphones got boring. We started @nothing to make them exciting again.
AI is changing everything. Software is easier to build but harder to distribute. That’s why hardware at scale matters more today than ever.
We’re approaching $1B in sales this year, and we’re just getting started!
Can a startup take on Apple and Samsung? @getpeid CEO and founder of @nothing joins @timkhiggins and @mims on the Bold Names podcast to talk about the future of the smartphone. 🎧 Listen: link.chtbl.com/WSJBoldNames
Christopher Mims 🤌 retweeted
Total factor productivity growth is slowing down, but we have more patents than ever. What gives? Aakash Kalyani argues that if you look at the actual text of the patents, they’re getting less creative — a change likely caused by falling population growth. 1/
Christopher Mims 🤌 retweeted
New: Across Silicon Valley, AI researchers are putting in 80-100 hour workweeks to keep pace with technological progress.
Several compared it to war, and one executive even joked that the new work schedule is "0-0-2," midnight to midnight with two hours off.
Who could possibly spend hundreds of billions on data centers? How about... everybody? OpenAI alone has proposed a total of 26 gigawatts of new AI infrastructure *in just the next 4 years*. Total cost: Around a *trillion* dollars.
/fin
wsj.com/tech/ai/openai-broad…
The scale of proposed AI infrastructure seems incomprehensible. But it's precisely this scale that makes it so attractive to the financial system. New forms of private debt for AI are being touted by e.g. Blackrock, -- Jamie Dimon calls it a "cockroach"
businessinsider.com/blackroc…
4/🧵
The other thing driving the AI bubble is vast amounts of capital (there's still $2.5 trillion cash on the sidelines) leftover from years of quantitative easing / economic expansion. Through equities, debt and VC, it's found its perfect match: insatiable demand for new AI infrastructure.
ft.com/content/a169703c-c4df…
3/🧵