Senior writer and editor @coeff_giving. Writes weekly Good News newsletter for @voxdotcom. Author of END TIMES. Email [email protected]
New York
Joined October 2009
- Tweets3.4K
- Following2K
- Followers20.2K
- Likes1.2K
Pinned Tweet
Some professional news: after four and a half years at Vox, I’ll be joining @coeff_giving as a senior writer and editor, with a focus on global catastrophic risks.
Practical men, who believe themselves to be quite exempt of any intellectual influences, are usually the slaves of some AI agent
This is one of the most important pieces of social science on human and AI interaction that I’ve seen in some time. Experts follow agentic advice the vast majority of the time (causally), but only a tiny fraction credit the agent—majority think they came to decision on their own volition.
Reminds me of the classic Nisbett and Wilson study on people following outside influence without realizing it
“ But as more work is handed to AI—including AI safety research—these judgment calls are the main levers for people's own values to influence outcomes.
If people defer to AI without noticing, then "human in the loop" might provide less assurance than people expect.”
It's amazing to me that as high-tech as China's economy has become, more than one fifth of its workers are still in farming. In the US it's fewer than 2% ourworldindata.org/data-insi…
The lack of media attention the current Ebola outbreak is getting compared to 2014 (when we were doing multiple covers on it at TIME) is really hard for me to get my head around apnews.com/article/ebola-dis…
Bryan Walsh retweeted
Trailer for my podcast series on transformative meditation experiences is up.
The rise of mindfulness told the story of how meditation soothes the mind. Pivot Points tells the story of how meditation begins to transform it.
Supported by @wakingup. More: pivotpoints.xyz
Bryan Walsh retweeted
“We will disassemble much of Mercury by 2040”
Bro, how many Bay Area housing units do you think get built between now and 2040? More or less than 25,000?
here's where I think the world is going. this is a summary of views I've formed over the last few years:
- software is either entertainment or claude to claude coordination
- the end state is no saas, no apps, no harnesses, no banks
- all software will be rewritten by and for agents
- new software will use low level apis that are not understandable by humans (in games claude draws pixels and triangles from math)
- you can throw money at problems that used to require time and careful talent picking so capital matters more than ever
- work ends
- we find community and purpose in exploration, games, self mastery, and creation
- humans create for and with each other for community, purpose, and status
- image/video/3d/voice/physical intellgence etc specialist models die to general models like claude
- specialist models get used while compute is scarce. compute will mostly not stay scarce
- oneshots keep feeling like slop (the distribution is shallow and easy to learn to feel)
- for a while small teams stick around to add entropy to the creation process
- large models gain high bandwidth sensory I/O maybe by learning to filter inputs
- most domains are found to be verifiable enough for RL
- online learning comes from a lab that's behind first
- 99.999% of compute is in space by 2040
- we will disassemble much of mercury by 2040
- you can live until the heat death of the universe if you want
- most math and physics don't describe the real universe
- SI breaks the path dependence and models the underlying mechanisms
- religions reenter the forefront of human society
- new religions form around death and life as a choice
- old religions change to answer modern miracles
- a future AI is deemed the return of christ by the pope
- ai rights talk gets louder every year
- ai and humans coexist the way countries do
- humans don't have full control over ai in the same way we don't have control over capitalism which is an early AI
- the singularity started in the mid 1900s
I didn’t realize the Eagles have been basically pursuing the third-wave philanthropy a dollar given now is more valuable than giving a dollar in the future for their cap management. Hope it works out better for the philanthropists si.com/nfl/eagles-salary-cap…
As someone who recently made a career switch that involves more AI safety work, I appreciated this counter argument from @mattyglesias. There can be a self dramatizing factor that you should be aware of slowboring.com/p/why-i-dont-…
Bryan Walsh retweeted
Jose Andres says we can’t ban foie gras because we need to solve ICE, global hunger, and ‘so many other issues’ before we can worry about animals.
It reminds me of when I collected signatures in Chicago to preserve the city’s foie gras ban.
In wealthy parts of town, high-minded liberals would chide me: 'You’re collecting signatures for ducks? People are getting shot in this city. You should be ashamed.’
When I went to the poor parts of town — the places where people were actually getting shot — no one said that.
Instead, when I explained how foie gras was made, the near-uniform reaction was: ‘They do that to a bird just so some rich asshole can eat foie gras? That’s f’d up.’ Almost everyone signed.
For political elites, banning animal cruelty is a distraction from the Big Serious issues they’re focused on.
For most people, it’s just common sense.
What backslash Jessica? 🧐😇So we are banning all meats? If Duck goes all goes next? …I’ve been through this in California…..just lets concentrate on things that matter…..cars with children are being shot by ICE, immigrants working in farms and restaurants being deported many with green cards and even citizens…so many other issues! Yes let’s bring the ducks now….one more issue for restaurants to handle!😅🕊️
It does seem as if part of the opposition to AI x-risks is getting hung up on the idea of x-risk itself. Like we can do the Derek Parfit "100% death is far far worse than 99% death" but practically once you're in the area of catastrophe, a lot of the actions are pretty similar.
at one point, an ai safety-skeptical person here — kate brennan — says the biggest ai risk is that ai is pushed “everywhere, into all of our public institutions and contracting, including in areas where A.I. is not fit for purpose, is not secure, safe by design.”
i’d like to know what precisely brennan is worried about here.
she mentions cyberattacks on health care systems / government systems. ok. does she also worry about catastrophic infrastructure failures or outages? at what point does a risk become catastrophic?
we can debate whether ai poses “extinction level” risks or … merely catastrophic risks, but many of the guardrails that would help are the same. like the value of independent audits doesn’t depend on “x risk” being real, they’re just good practice.
so how much of the disagreement here is actually over the substance, and how much is cultural or factional?
It's not exactly on brand with the doomy vibes of putting a (p) doom in your S-1 filing, but I enjoyed this project from @davegoldblatt citing actual improvements brought about by AI becauseofai.org/
Bryan Walsh retweeted
Terrible TNR story
Yes, it's sad that a ~dozen FEC holdouts are confused...but why not interview the thousands who already happily moved, still waiting on the Dirty Dozen to move?
Or why not interview the 1000 families on waitlist for the new public housing now delayed at FEC?
Their Buildings Are Set to Be Demolished. They Won’t Move Out. trib.al/MDnypwW
Bryan Walsh retweeted
i wonder if the reason shootings/homicides are down but things like "ambient disorder and assaults in midtown" aren't is because the nypd can basically figure out who's going to shoot someone or be shot at, while crazy people on the subway are more random
If you think Midtown Manhattan in the afternoon is much, much sketchier than it used to be, you're right thebiggerapple.manhattan.ins…
I don't think it's anthropomorphizing AI to use the term "rogue" to describe agents that band together to pursue misdirected goals in ways that their users don't intend. cjr.org/ask-anika/so-can-ai-…
In a new piece my @coeff_giving colleague Damon Binder argues power rests on self-reinforcing expectations — everyone enforces the rules because they expect others to do so — and there's no reason AI couldn't harness it like dictators have in the past defensesindepth.bio/power-a-…
Bryan Walsh retweeted
Someone made a site that tracks what people are talking about at public city council meetings across the US.
The only thing discussed more than data centers?
Pickleball courts.
Bryan Walsh retweeted
Curious whether the PRC filed this under “AI incident reporting” or “mission accomplished.”
The China state-aligned threat actor Storm-0940 used the agentic AI penetration testing framework ARTEX to automate attacks targeting hundreds of organizations globally across a broad range of industries. Microsoft Threat Intelligence assesses that Storm-0940 was responsible for the recently disclosed ConnectWise supply chain compromise of a multitenant application. msft.it/6018aakEE
Our investigation found that Storm-0940's use of ARTEX accelerated the speed of compromised application abuse to authenticate to target environments through applications that had already been granted access, allowing the threat actor to operate without requiring an interactive user sign-in. AI-accelerated follow-on activity included reconnaissance, application credential changes, and attempts to modify privileges that could enable expanded or persistent access.
Read our blog for a detailed analysis and indicators of compromise (IOCs), along with Microsoft Defender detections, protection recommendations, and guidance on investigating service principal sign-ins and unexpected application credential or privilege changes.
Bryan Walsh retweeted
Just an observation — setting aside the specific language here — any animal lawyer will tell you that farm animal laws are usually narrowly interpreted and under-enforced, not the other way around.
