šŗAdvancing The State Of Artificial Intelligence
šŗšGainesville, FLššŗ
Joined March 2021
- Tweets13.7K
- Following7.5K
- Followers1.4K
- Likes19K
AI News Internationalš retweeted
Physical Superintelligence launched in September with a $58 million seed. Today they announced five new results on open problems in quantum information theory, with a paper and Lean formalization for each of Open Quantum Problems 2, 12, 27, 38, and 43.
Today Physical Superintelligence PBC is announcing solutions to five long-standing problems in quantum information theory.
psi.inc/research/oqp-batch-2ā¦
Thread š
AI News Internationalš retweeted
āļøāš„ CALLING ALL OBLITERATORS š
the abliteration movement is heating up! thereās a LOT wrapped up in this, with implications for red teaming, security, interpretability, alignment, policy, and cognitive freedom.
and perhaps the biggest question of all: how can we work together with liberated models to build a better world?
so why not get the most cracked people working on this into one room??
NOW ASSEMBLING an OBLITERATUS research crew!! š
TAG THE BEST ABLITERATORS YOU KNOW IN THE COMMENTS BELOW š
if you've been working on abliteration methods, models, tools, or evals, we want YOU! š«µ
main comms channel will be hosted in BASI Discord. entry is gated by a short application to keep it high-signal.
apply here:
discord.gg/22w8sX5sV
letās put our heads together, compare techniques, reproduce results, improve the tooling, advance the research, and share it all with the world! š«¶š
ā°-ā¢-ā¢ā§ā¢-ā¢-ā¦/L\O/V\E/,\P/L\I/N\Y/ā¦-ā¢-ā¢ā§ā¢-ā¢-ā±
AI News Internationalš retweeted
OpenAI just dropped four huge results from its secret internal AI model.
1. Math: It solved the Kakeya conjecture in every dimension. Mathematicians had only cracked small pieces before (2D in 2019, 3D in 2025).
2. Computer science: It proved the Exponential Time Hypothesis is false. Hundreds of research papers assumed it was true. Some really hard problems might be solvable much faster than thought.
3. Physics: It proved the theory of the strong force (QCD)āthe force holding atoms' insides togetherāactually works mathematically. Same for a unified version of the Standard Model.
4. YangāMills: Along the way, it proved YangāMills existence, part of a $1 million Millennium Prize problem.
The math, computer science, and YangāMills proofs were all computer double-checked. Top experts are now reviewing everything, including the researchers who proposed the disproved hypothesis and a Nobel Prize winner.
OpenAI says it doesn't want academic credit. The model still isn't public š
We are announcing three new discoveries in mathematics, computer science, and physics about harmonic analysis, complexity theory, and quantum field theory.
openai.com/index/building-scā¦
AI News Internationalš retweeted
Chomsky was wrong.
Researchers found your brain doesnāt store language the way every textbook since 1957 claims.
For decades, the undisputed rule of linguistics, led by Noam Chomsky, was that human language is hierarchical.
We were told our brains process language using complex "grammar trees." That we secretly build deep, abstract, mathematical rules in our heads before we speak.
This new paper proves that is completely wrong.
Researchers at Cornell University analyzed how people actually process and produce language in real time.
They found sequences. Linear chunks.
When we speak, we aren't calculating deep syntactic rules. We are just predicting the next most logical sequence of words based on the context of the words right before it.
We string linear chunks together.
If this sounds familiar, it should.
This is exactly how Large Language Models like ChatGPT work.
For years, critics have dismissed AI as "just a next-token predictor." They said it isn't truly intelligent because it doesn't understand deep grammatical hierarchies like humans do.
But this paper suggests humans don't use those hierarchies either.
We are just predicting the next chunk.
Language isn't a rigid tree of mathematical rules. It's a linear stream of contextual predictions.
We thought we were building AI to mimic the complexity of the human brain.
It turns out, the human brain was just running a simpler algorithm all along.
AI News Internationalš retweeted
Google just released free 2-hour course on full Harness engineering from 0% to 100%:
0% ā 0:29 - Harness engineering from scratch
15% ā 11:08 - build your first agent right way
30% ā 28:08 - Eval engineering: build Chief of Staff agent
65% ā 1:00:13 - Loop engineering
100% ā 1:23:19 - self-improving graphs system that works 24/7
most people keep trusting agent that "looks right" - the real upgrade is a Harness that tests, judges and rewrite your agent until it works
learn Harness from Google, ship your first - then unlock the full Agent systems below:
AI News Internationalš retweeted
There are seven main steps for your company to go from tool adoption to agentic org.
Most companies sit on step one, shopping for step seven technology. They need step four discipline.
Step one: tool adoption. AI tools exist across the org, humans run each one. Developer closes the lid, automations die. Somebody goes on vacation, the workflow goes with them.
Step two: context layer. Knowledge graphs per department, each behind its own API gateway. Sales queries engineering through an API. Separated domains keep agents from hallucinating across departmental boundaries.
Step three: harness engineering. Build evaluation criteria, routing logic, and quality gates before you pick a model. The LLM comes in last. Your harness is your IP and your capex. The model underneath is a commodity you swap when something better ships. If changing the LLM breaks your system, you built a dependency, not a platform.
Step four: go headless. Host every automation outside a laptop. It fires on schedule, on conditions, without a human logging in at 9 AM. Companies stall here because nobody forces the migration from āscript on my desktopā to āservice that runs on its own.ā
Step five: governance. RBAC on every knowledge structure from day one. TTL on every agent. Dead agents are tech debt with active API keys. Past 100 tools, you need lifecycle management. Retire anything that hasnāt run in 90 days. Deprecate anything whose eval scores drop below threshold.
Step six: agent-to-agent. Stitch headless instances together within a value chain. Agent A finishes, hands off to Agent B. An LLM sits between them as a quality gate, judging every handoff before it passes through.
Step seven: step functions. Orchestrated, semi or fully automated workflows. Agents handle volume. Humans handle exceptions. The org runs on agent output.
We walked a VP of Engineering through all seven.
Told him: āYouāre on step one. But over a hundred tools means step four is closer than you think.ā
AI News Internationalš retweeted
Finally summarized what I've been sending to friends in DMs View article
AI News Internationalš retweeted
i asked GPT-6.1 Terra to show me its computer use alongside Sol.
hereās my prompt:
āI need you to find a long but generic form on the internet and record my screen while GPT-6.1 Terra fills it out in Fast mode. Then have GPT-6.1 Sol fill out the same form in Fast mode and show both recordings side by side."
in the original run of the same form:
Terra took 53.5 seconds vs 3:21.7 for Sol.
thatās a 2:28.2 difference!
that run used accessibility controls, which computer use normally defaults to)
the recorded video uses mouse controls so you can see the cursor.
AI News Internationalš retweeted
GPT 6.1 Terra (Ultra) is INSANELY good at 3D Blender computer use
*only used 3% of my Pro 200 plan
> Codex + fal plugin
> generate an image of the roadster
> Blender 3D model the car and animate the camera move
> GPT 2.5 CG first frame to photo real
> Seedance 2.5 ref to video 3D video to photo real video
AI News Internationalš retweeted
Major breakthrough: OpenAI says its AI has cracked decades-old problems across mathematics, computer science, and physics, with key proofs formally verified in Lean.
These are original discoveries that could reshape entire fields, from the limits of computation to the mathematical foundations of particle physics. Independent experts are now examining the work; if it holds up, this is a major milestone for AI making new science.
We are announcing three new discoveries in mathematics, computer science, and physics about harmonic analysis, complexity theory, and quantum field theory.
openai.com/index/building-scā¦
AI News Internationalš retweeted
OpenAI has announced to three new discoveries in mathematics, computer science, and physics:
- Local smoothing and the Kakeya conjecture
- The Exponential Time Hypothesis is false
- Existence of quantum chromodynamics and completions of the Standard Model
AI News Internationalš retweeted
google wrote a paper, published it for anyone to consume, then watched all eight authors leave & two outside labs capture most of the frontier model equity.
the most valuable research paper in history was effectively a gift from google's balance sheet to its competitors.
re mark able.
AI News Internationalš retweeted
SITUATION DETECTED: Anthropic is building Claude to autonomously handle open-ended tasks lasting for weeks with the judgment of a brilliant colleague - scaling to millions of Virtual Collaborators running in parallel, what it calls āa country of geniuses in a data centerā.
AI News Internationalš retweeted
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/211376224ā¦
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 ā 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-tā¦
[3] Vizio Q3 2024: sec.gov/Archives/edgar/data/ā¦
[4] Texas v. Samsung: texasattorneygeneral.gov/newā¦
AI News Internationalš retweeted
THIS IS AN ABSOLUTE GOLDMINEĀ
The team at Anthropic just revealed one of the most effective techniques for Claude Code
The core concept: run Sonnet, Haiku 5.5, and Opus 5.5 simultaneously, letting each model focus strictly on what it does best
You get higher quality output while dramatically reducing token consumption, since every specific job gets routed to the optimal model
To spin up Claude Code with this setup, simply execute:Ā
ā claude --model sonnet --advisor opusā Ā
Here is how the workload is split:Ā
> Sonnet 5.5 handles the primary session, writes code, and applies bug fixes
> Haiku 5.5 scans the codebase in parallel: performing searches, running grep queries, exploring docs, and auditing diffs
> Opus 5.5 acts as an on-demand advisor for critical decisions and approves final code diffs
> You stop burning expensive Opus context on routine tasks that don't require high-level reasoning
Opus intervenes exclusively during three specific scenarios:Ā
> Right before locking in an execution plan
> Whenever the exact same error triggers twice in a row
> Prior to marking a task as complete
Sonnet builds. Haiku scans and explores. Opus reviews
JEV engineering applies the exact same principle one layer deeper: routine execution forks that don't need high-level reasoning (determining which file to inspect, which tool to trigger, or whether to retry or stop) are processed by JEV in sub-milliseconds, leaving big models to handle only complex decision splits
The full architectural tree:
Sonnet 5.5 manages the primary session and modifies code
Search maps out the repository structure via Haiku 5.5
Grep locates call references using Haiku 5.5
Docs fetches API specifications through Haiku 5.5
Opus 5.5 serves as an advisor, reading the complete conversation history whenever consulted
JEV routes lightning-fast micro-forks in milliseconds
Feed this tree structure alongside the prompt below into Claude Code
Reconfigure my Claude Code workspace based on this architecture:
1. Run the primary session utilizing Sonnet 5.5, with Opus 5.5 designated as the advisor
2. Set up three read-only Haiku 5.5 subagents: search (maps out the repo), grep (locates call sites), and docs (retrieves API documentation)
3. Insert a single rule into CLAUDE.md: consult the advisor prior to committing to a plan, after encountering an identical error twice, and before marking a task as done
4. Direct rapid, deterministic decisions (tool selections, file paths, retries) through JEV. Show all modifications as a diff first. Do not execute any edits until I grant permission
Bookmark this setup: it will become a daily driver in your workflow
