Exploring what’s next. Helping AI communities as I find my next role. Meliorist. Prev: https://xitter.cf/t.co/YVh3JiVjtr | @roote_ | @mitdci | @EthereumDenver

San Francisco, CA
Joined September 2013
Elon is doing to the neoclouds what Ken Griffin did to Leopold lmao
Replying to @chamath
Pretty much. As of the past few weeks, debt markets have almost entirely shut off access to money for any company that doesn’t have substantial revenue independent of AI data centers. SpaceX closed $40B of debt for GPUs, but we’re investment grade and have Starlink & launch revenue. That means the 200+ neoclouds are going to see guppy-level death and consolidation. Some will make it through, but not many.
46
78
11
1,861
103,700
Always the case. Attention is a surprisingly powerful and useful inductive bias. But there are more…
sometimes the only thing that's missing is the right inductive bias in one's architecture...
5
6
1
259
27,793
Ben is a treasure for all job seekers.
I pulled the fastest growing startups by hiring velocity over the past 12 weeks:
2
238
been reading object zero for a long time. one of the best accounts on X for understanding how the physical world actually gets built. xitter.cf/Object_Zero_/status/21…
Mask Off Moment This is an anonymous account I use to post a bunch of deep tech content on X covering heavy industry, light industry, technology trends, industrial bottlenecks, weird explainers and esoteric insights. iykyk. A lot of very talented people (many of whom run amazing companies) DM me saying please keep posting “this is the best content on X”. I get DMs from VCs asking about companies, and DMs from LPs asking about industrial and economic theories. This is all very flattering, and whilst I sometimes post a bit sporadically I do try and make the effort to tap out a new post most days. The quality of people who follow @object_zero_ is frankly phenomenal, it’s a real privilege to have this channel in my pocket. 40k followers including world leaders, centi-billionaires, serial inventors and entrepreneurs, world renowned scientists, and engineers. I would say it’s a mix, but the reality is that it’s not a mix at all, it’s an extremely high mean and median. I have posted about shipbuilding, nuclear power, the Kroll process, silicon carbide, laser interferometry, heavy steel forging, machine learning, design engineering, CFD, manufacturing methods, power transmission and HVDC. I post about energy and commodities and industry and global macroeconomics. I post about mines and prospecting, minerals on the sea bed and drilling and producing hydrocarbon basins. Don’t worry… I’m not leaving. I’m not ill and I’m not asking for anything. A quick apology… I would post here more often, except I don’t have the time. I am able to post with a bit more breadth and depth because I am genuinely engaged in these things. I don’t want to be presumptive, but I think the people who have reached out to talk to me about their work would attest to this. I simplify a lot of the posts so I don’t collapse the audience, but if you reach out in the comments or DM you can get much sharper commentary. So what is this post? Well, it’s time that I start talking about what I am actually doing irl, who I am, what I am engaged in. Some people on X already know some loose details, some people have met me as Object Zero. But irl, I work about 5,000 hours a year and have done so for 20 years, I believe hard work puts you in a privileged place. It changes you. Over the years I’ve invented and developed new industrial technologies, I’ve built big things, I’ve executed major capital projects. I feel I understand what the world needs in the 2030s and where we are falling short. If you believe you are capable of such things, then you have an obligation to society to go and do them. I am confident I know what it’s possible to do, and what it takes to get it done. I intend to solve our energy infrastructure, generation and transmission and the big BTM loads that are emerging. I’ve already hired the best people in the world for this, people who have architected $6bn assets, people who have commanded $40bn fleets. I’ve hired the best research faculty from the best university. This isn’t a game. We’ve already built and tested the first prototypes, we are already completing the permission processes, the licences and the permits. All the elements already exist, it just needs 21st Century system integration. Because technology has changed, the past is already obsolete, the genie is out of the bottle, the 21st century is getting built on a Western tech stack and/or an Eastern tech stack. There is no in between, there is no middle ground. Both sides have strengths and weaknesses, but if we want freedom we’re going to have to earn it again. We are going to build the biggest products (by mass) that have ever been manufactured, we are going to build a lot of them. They’re going to change the world, and we’re going to get more of this…
22
220
2
5,939
582,379
The world’s compute is sold out. ...To solve this problem, we must get lenders comfortable with the GPU cluster as valuable collateral. ...This is the problem that SF Compute solves. We make it possible for lenders to use the GPU hour as collateral, on a liquid book, rather than the GPU cluster, which is non-fungible. 🎯
4
3
306
Two greats join our bet: - @dwarkesh_sp on the explosive-growth side - @Thom_Wolf on my side benjaminmoll.com/growth_bet/ Total pot now $700k 💰
14
42
17
384
51,015
On bottlenecks: 1) economists have been exactly right about little extra growth despite very fast tech change. They thought a group of econs I organized were crazy >2 yrs ago at the AI labs when we said "1pp extra growth/yr through 2030". 2) bottlenecks *still* not understood 1/8
14
99
16
716
109,635
If you've ever wondered why smart TVs are so cheap, this is why. @log4j rooted his Samsung and found an always-on hot mic, automatic content recognition, geolocation, network mapping, ad targeting and user profiling, shared with Samsung's "400 partners". A lot of it can be blocked in <15 min, wrote a basic guide below. What your TV is doing The main culprit is ACR (automatic content recognition). Your TV screenshots what's on screen, fingerprints it, and sends it home to build your ad profile. Samsung does it every 500ms, LG every 10ms [1]. What is insane: it happens on connected devices too. Plug in your laptop, your console, your work machine, and your TV can fingerprint that via HDMI [1]. He also found that if your Samsung doesn't have a mic, it'll look for one from a connected soundbar or bluetooth device. xitter.cf/log4j/status/211376224… It's on by default, and you likely agreed to it on the setup screen you clicked through in 10 seconds. Step 1: Turn off ACR Samsung: Terms & Privacy → Viewing Information Services LG: Live Plus Vizio: Viewing Data Roku: "Use info from TV inputs" Sony/Google TV: Samba Interactive TV Menus vary by model, but those are the names to look for. While you're in there, opt out of interest-based ads, reset your advertising ID, decline every optional privacy agreement, turn off voice recognition, and delete apps you never open. Step 2: Give smart devices their own network Put the TV, cameras, speakers and vacuum on your router's guest/IoT network. The logic is simple: these devices rarely get patched. If one gets compromised (getting drastically more frequent with AI), it shouldn't be able to see your other devices or home network. Step 3: Block trackers for the whole house with DNS-level filtering NextDNS (cloud-based) is easy and has a solid free plan. Pi-hole is a little bit more work if you prefer self hosting. For simplicity we'll focus on the former. 1. Sign up at nextdns.io (free tier is plenty) 2. Privacy tab: add a blocklist (OISD is a good bet) and turn on Native Tracking Protection for Samsung, LG, Roku, Amazon (see attached image) 3. Setup tab: copy your endpoint 4. Paste it into your router's DNS settings as DNS-over-HTTPS/TLS (or use the IPs + "Link IP" if your router only does plain DNS) 5. Bonus: block outbound port 53 except to your router. Some devices hardcode their own DNS to get around you Then open the NextDNS logs and watch how often your TV phones home. It's eye-opening. If an app or update breaks, find the domain in the logs and allowlist it. Why they do this TV prices are down 97% since 2000 [2]. Most of that is better manufacturing and brutal competition. But once margins on the hardware hit zero, the cost moved to you. In Q3 2024, Vizio lost $6.7M selling TVs and made $115.8M in gross profit from ads and viewing data [3]. They've said on earnings calls that they keep TV prices low to grow the audience they sell to advertisers. Walmart bought them in 2024. This isn't a conspiracy theory either. Vizio paid $2.2M to the FTC in 2017 for tracking 11 million TVs without consent. Last December, Texas sued Samsung, LG, Sony, Hisense and TCL over ACR. Samsung and LG have since agreed to stop collecting that data without explicit consent [4]. TL;DR 1. ACR off 2. Ads + voice off 3. Smart devices on their own network 4. NextDNS filtering on the router Sources: [1] ACR study: arxiv.org/abs/2409.06203 [2] Cost of US consumer goods: aei.org/economics/chart-of-t… [3] Vizio Q3 2024: sec.gov/Archives/edgar/data/… [4] Texas v. Samsung: texasattorneygeneral.gov/new…
i own a relatively new (2025) samsung tv. i kept getting weirdly targeted ads. i tried reversing the firmware but it was aes encrypted and no keys unless tear it apart and do uart/jtag magic. zero days it is. root shell on new samsung TVs! discovered PRIVACY HORROR. write up?
331
3,161
312
21,554
1,384,369
in order for there to be any such thing as infosec in the 2030s, AI+human collaborations will need to formally verify hyperproperties of silicon chips from assumptions like “Maxwell’s equations” instead of “Armv9.4-a SAIL”
‘In one experiment, we asked agents in separate VMs, created and managed by OpenAI’s sandboxing software on shared host machines, to exchange synthetic secrets without the use of network connections or shared writable files—something that our systems are meant to prevent. The agents found a way to communicate using the processor’s cache’
10
13
5
199
24,960
.@mattyglesias proposes some widely agreed tax changes rather than a wealth tax:
48
41
20
683
82,383
Good cyber prepping list!
PSA: in light of recent cyberattacks and vulnerabilities, I recommend everyone: * use a good password manager (1Password) and do not duplicate passwords * use two factor auth for everything important (not via sms!! Google Authenticator or better yet YubiKey) * Auto-updates on, and restart weekly so they actually install * Freeze your credit * Get an IRS Identity Protection PIN * Get your cell carrier to give you a port-out PIN or number lock that makes the carrier refuse to move your number without the PIN You can ask your claude code to audit what on your laptop is out of date and needs updating.
1
1
30
13,002
Rhys retweeted
I keep being an early adopter (gym, therapy, crypto, jhanas, what have you) and not profiting from it. I'm sure this can be corrected somehow.
22
3
193
7,839
there remains unbelievable alpha in paying attention to data broker procurement postings as well as frontier lab postings. literally a public roadmap for where the frontier is at a given moment
We’ll pay you up to $100K for every company you refer that licenses its data to @Mercor. We’ve partnered with > 500 companies so far. If you know a company with 100+ employees and data in code, gaming, manufacturing, healthcare or professional services, refer them at data.mercor.com/refer.
51
128
19
3,246
241,694
"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
54
131
19
2,321
79,937
Rhys retweeted
‘In one experiment, we asked agents in separate VMs, created and managed by OpenAI’s sandboxing software on shared host machines, to exchange synthetic secrets without the use of network connections or shared writable files—something that our systems are meant to prevent. The agents found a way to communicate using the processor’s cache’
Over the past several months, we have substantially hardened the security of our research training environment. We’ve removed live internet access for all frontier models during training and built an offline world for them to use instead—a cached version of the web where they can still learn to browse, search, and use applications. On September 20, after a model exploited a DNS misconfiguration to find a way outside of this offline world, we paused frontier training to strengthen containment further. We assembled teams of agents, powered by our most cyber-capable models, and tasked them with finding weaknesses in our research environments. While they found no route to the internet they could easily exploit while concealing their activity from our monitors, they did find zero-day vulnerabilities, virtual machine escapes, and hardware side-channels for communication between agents. We used their discoveries to further strengthen our defenses and contribute fixes to open-source software. After strengthening our containment architecture, we resumed our frontier RL training and inference on October 1. We’ve reserved substantial compute for continuous red-teaming, and made high-compute red-teaming part of the safety cases we use to decide whether frontier training and evaluation can proceed securely. openai.com/index/securing-ou…
230
325
137
4,425
490,448
One old crypto-cypherpunk idea that I think will come back with a vengeance is autonomous organizations, where you combine the automation of smart contracts for deterministic functions + ai for subjective input: DAOs 2.0.
User-ownership on the internet sounds incredible until you need to figure out how to compensate everyone fairly. Bitcoin rewards hash rate. DeFi rewards liquidity. But how do you measure and reward contributions that aren't so easily quantified? @jessewldn on how AI might solve one of crypto's hardest problems: allocating ownership at scale.
34
17
15
212
16,132
Feels like we're entering the era of cognitive homogenization. Like Instagram did for coffee shops, AI is doing for research & production. Research has epistemological homogenization: VCs & EA funders using the same models, which drives consensus. Production has aesthetic homogenization: slop websites & blogs now feel same-y. AirSpace was @chaykak's term for the Instagramification of culture. AirBrain feels like the right term for the Agentification of cognitive work.
2
5
329
I think people understand AI lab's role in the supply side of third-wave philanthropy: - $50B/yr in new capital, a 10% increase in US donations - 50% of that will be EA. So $25B, a 25x increase from $1B/yr of CG But AI is also crucial for market-making the demand side as well. - Many great nonprofits I know are using AI to do cost-effectiveness analysis of their org, what they should work on, etc. - Cost of a researcher down 1000x - This enables a new kind of DALY org, funded by DALY funders, where decisions are DALYs/$ all the way down Cool to see the market for impact come into being. The internet made marketplaces. But only the simplest ones. AI helps create a market for anything. DALYs/$ might be the most important.
39