Context architecture and memory curation for high-performance Papers and PKM Data Neuroscience/AI
Data center AI
Joined November 2015
- Tweets8.6K
- Following809
- Followers762
- Likes15.8K
AgentX retweeted
The kids who were told to learn code who then are losing their jobs to Ai are going to code Ai to take down companies that sold them the lie to learn code and then fired them.
It is becoming f*cking dangerous
A 23-year-old dev from China just ran 300 Claude Code agents in parallel and hooked them to a verification layer so none of them could bullshit him
He opens the live dashboard:
- 300 agents writing and checking at the same time
- Every output verified against the source in ~100 ms
- Three passes over 100 company profiles → zero errors left
- Everything dumped straight into an Obsidian vault
The trick is the split: one model does the work, a cheaper one decides if the claim is actually true. The loop doesn’t stop until everything checks out
An analyst team would bill weeks for this. He got it in one run
Same kind of Second Brain + swarm setup that keeps circulating. Bookmark the full breakdown
AgentX retweeted
Wow. I woke up to an impressive number of upset people who apparently don’t understand how LLMs work and can’t cope well with me pointing it out.
An LLM is a statistical next-token prediction model. That’s how it works. It’s an extraordinarily sophisticated calculator performing math you could theoretically do by hand (if you were sufficiently masochistic).
Math doesn’t know what it’s calculating. Pi doesn’t spend its evenings contemplating the nature of circles, and c doesn’t ponder the relationship between matter and energy.
That doesn’t make the results any less accurate, impressive, or useful.
Between inference requests, an LLM isn’t sitting around thinking. It isn’t reconsidering its answers, wondering whether it made a mistake, contemplating its existence, or questioning whether it’s conscious. Its weights remain fixed (it doesn’t learn), and computation resumes when it receives input.
We know this because we designed these systems. The architectures, mathematics, implementations, and research are extensively documented. You can study them yourself.
Of course, some people immediately lean into comparisons of biological intelligence or philosophical debates over consciousness because that is what they know.
But biological intelligence and artificial neural networks are fundamentally different systems. Similar outputs don’t establish equivalent internal experiences.
I experience consciousness directly. I infer that other humans do too because we share biological structures and processes. We don’t have equivalent evidence for an LLM, and nothing about its known operation gives evidence of conscious experience.
Which brings me back to my original point.
To me, understanding is conscious comprehension. It means being aware of information, recognizing its meaning, and experiencing that comprehension. If I wasn’t conscious I could not understand.
We can perform actions unconsciously. We can produce correct results without knowing why they’re correct. But I distinguish those capabilities from consciously understanding what we’re doing.
An LLM can solve complex problems, explain difficult concepts, and produce remarkably accurate information. I give it full credit for those capabilities.
But producing an answer that we rationalize as understanding isn’t the same as experiencing understanding.
You don’t have to agree with my definitions. We can debate functional understanding versus conscious comprehension all day, but that won’t establish whether an LLM actually experiences anything.
The right debate is this…
Could machines exhibit something different from human consciousness, something measurable, meaningful, and worth studying?
I think so, absolutely!
I’m entirely open to discussing “machine consciousness” in terms of observable, testable computational properties. We built these systems, we can instrument them, examine their internal mechanisms, and investigate what they’re capable of.
That seems far more productive than endlessly arguing over whether a machine experiences life the way humans do.
So here’s my question:
How would you define machine consciousness, and more importantly, how would you measure it?
AgentX retweeted
Dusting off this tweet from April because this gap in shared understanding of LLM capability is *widening*. It's now less "two groups of people speaking past each other" and more a sharp funnel.
- Napkin math somewhere around 6B people (~75% of the population) have barely come in contact with LLMs at all.
- Around 1-2B (~20%) are casual and infrequent users of free-tier ChatGPT-like products. This group sees derpy chatbots and treats them a bit like a better Google search, a writing aid, or etc. Maybe an agent tries to book you a flight. I have non-tech friends who (reasonably, imo) say they have not much use for it at all. Even many professionals outside of math&code are in this tier. For example, execs and many other functions spend a lot of their time talking to other people, so while they understand what is happening intellectually, it is still second-hand and a bit abstract.
- Now we get to professional use of frontier-grade LLMs in math&code. Somewhere around 20M people (0.2%) see first-hand that large, complex projects that used to take them weeks/months can now be completed by agents with a prompt. Building apps, copying apps, translating apps, decompiling apps from binaries... This has all happened very quickly and recently - less than 1 year ago, I was writing code manually by hand, typing memorized computer code commands into a code editor character by character, occasionally pressing Tab to autocomplete a little chunk of code.
- And finally we get to the ~5,000 people (~0.00006%) with access to frontier-grade systems internally. The external world has seen the preview. It looks like swarms of thousands of agents collaborating over weeks on software mega projects: minting zero days, running cyber attacks and defenses at machine speeds, discovering new science, advancing the frontier of mathematics. Things that would have taken top professionals in the industry years of work. Meanwhile, human review and comprehension are starting to fall behind. For example, people are still involved in the "archeology" of the OpenAI-HF incident from many months ago. Mathematicians may be poring over the 722 manuscripts on frontier mathematics for a while.
The funnel is driven by a combination of factors:
1. The impact scales with ambition, problem size, and horizon. A question with a paragraph answer barely stresses the system. You need a reservoir of big, difficult problems that you really care about. This aspect drives the consumer / professional dimension of the funnel.
2. The jaggedness of the system (which I have written about a lot separately). Capability peaks in domains that are digital, verifiable and economically valuable. This is because LLM capability emerges from reinforcement learning on verifiable rewards on a curated environment mixture driven by revenue potential. This aspect primarily drives the area (e.g. math&code) dimension of the funnel.
3. Access. Free-tier, paid-tier, internal.
So this is the weirdness of the moment. The general public has mostly not interacted with these systems. When they have, it looks like a derpy chatbot. The majority of professionals still see only a modest uplift. And a small sliver of professionals are experiencing the vertigo of the curve going vertical. And it is all happening at the same time.
Judging by my tl there is a growing gap in understanding of AI capability.
The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code.
But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along.
So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions.
TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.
I love this ... Is amazing 😍
What if,
for any scientific claim,
you can search 100M+ papers to find the strongest evidence for and against it
⚡️in 1 second?
That’s now possible with Evidence API! 🚀
paperclip.gxl.ai/evidence-ap…
For example, does drinking coffee reduce diabetes risk?
AgentX retweeted
🚨 The dead internet is no longer a theory.
China’s Manus AI, the autonomous agent built by Butterfly Effect, can run dozens of social media accounts simultaneously; posting, replying, and engaging around the clock with no human in the loop.
A single dashboard can orchestrate 50 profiles at once, turning one operator into a persistent online presence.
The footage is a wall of phone screens all moving at once. Most people scrolling will never know which accounts are real and which ones are just machines talking to each other. If this is already possible, how much of your feed is still human?
AgentX retweeted
We'll be spending a lot more time trying to understand the outputs of language models. A few thoughts, tips & tricks:
Writing. Something I've had success with: Ask your LLM to explain something in ASD-STE100, it's a controlled language specification originally developed for aerospace maintenance documentation. LLMs well-versed in this language and it comes with heavy constraints on clean writing style that I often find a lot more readable. Sometimes I've tried to soften it a bit e.g. ask for "80% of the way to ASD-STE100" because the spec is quite stringent. But even better:
Diagrams / images. Instead of writing, ask your LLM to create a diagram. These can be a lot easier to process, parse, and understand. But even better:
Web pages. Ask for output "in HTML" to get a beautiful, interactive webpage. LLMs are getting really good at frontend and can create beautiful experiences, animations, etc. But even better:
Explainer videos. The output format I am most bullish on is fully custom / bespoke explainer videos generated on any arbitrary topic. Experiment with things like "Create a 3b1b style video explainer on X. Use my ElevenLabs API key for audio narration". (you'd need an API key for the latter or you can ask your LLM to find you decent free alternatives that use your local compute). This is actually starting to work!
In summary:
- As LLMs get better, they will do more and more of the legwork autonomously, and a lot more of our work will rise up the abstractions into oversight and understanding.
- Luckily, LLMs can help here too because as intelligence and code are increasingly abundant, you can ask for large, custom, discardable software artifacts (e.g. web apps, video explainers) that would have never made sense to create before. Push the boundaries here and you'll be surprised.
AgentX retweeted
Read the full study in iScience: cell.com/iscience/fulltext/S…
AgentX retweeted
10 NeurIPS 2026 papers in robotics / embodied AI worth watching:
1) Embodied-R1.5
one model for physical reasoning, grounding, planning + robot control
arxiv.org/abs/2606.11324
2) LaST-R1
reinforcing the latent physical reasoning inside a VLA
arxiv.org/abs/2604.28192
3) LessMimic
humanoid interaction using local geometry instead of continuously copying human motion
arxiv.org/abs/2602.21723
4) Point4D
long-range 4D motion reconstruction through occlusions
arxiv.org/abs/2609.09145
5) Grasp-Then-Plan with Failure Attribution
figure out whether the grasp failed or the motion policy failed
arxiv.org/abs/2606.03385
6) OxyGen
shared KV caches for much faster multi-task VLA inference
arxiv.org/abs/2603.14371
7) PACE
help robots understand how far through a long-horizon task they already are
zhuochenn.github.io/
8) TDBench
tests whether VLMs really understand top-down / bird’s-eye scenes
arxiv.org/abs/2504.03748
9) Where to Attend / PaPE
positional encoding designed around visual geometry instead of 1D language
arxiv.org/abs/2602.01418
10) H-GenPO
hierarchical generative policy optimization for Physical AI
sunghoonim.github.io/
really interesting batch across robot reasoning, humanoids, VLA post-training, 3D perception, long-horizon manipulation and efficient inference.
AgentX retweeted
Apparently all of these “AI is escaping containment” stories are actually just AI researchers not knowing jack squat about the absolutely most basic security practices.
AgentX retweeted
Scientists mapped what happens inside immune cells when people exercise regularly. The results explain why physically active people resist infection better and show how the body remembers physical activity at the molecular level.
Researchers at BGI Research analyzed blood samples from 86 young adults: 40 who exercised at least 150 minutes per week, and 46 who were completely sedentary. They profiled over 1.3 million immune cells to see how exercise changes them.
The study reveals regular exercise creates three coordinated changes: enhanced energy use, strengthened defenses against cell damage, and immune cells reprogrammed to better detect threats.
Burning fat, building defenses
Exercisers showed clear signs their bodies were burning more fat for fuel. Multiple markers of fat breakdown were elevated, including molecules that transport fats into the cellular power plants where they get converted to energy.
At the same time, exercisers stored less fat as triglycerides. The body shifts from storing fat to using it.
Molecules that protect cells from damage were elevated across the board. Betaine, a compound recently shown to extend lifespan in animals by mimicking exercise effects, was significantly higher in exercisers. Other protective molecules that neutralize harmful byproducts of metabolism were also elevated.
Exercise also changed molecules produced by gut bacteria. One compound that strengthens intestinal barriers and reduces inflammation was elevated in exercisers.
Immune cells remember exercise
The most striking finding: exercise reprograms immune cells by changing which genes are easily accessible.
Your DNA is tightly wound around proteins. Some sections are packed loosely and easy to access, others are wound tight and harder to reach. Think of it like organizing a toolbox: frequently used tools get moved to the top drawer where you can grab them immediately.
In exercisers, immune cells reorganize their DNA so critical defense genes sit in the top drawer, ready to activate the moment a threat appears.
Monocytes, white blood cells that patrol your bloodstream looking for invaders, showed the clearest changes. In exercisers, genes responsible for detecting and displaying foreign molecules became more accessible. These cells were better equipped to sound the alarm when something dangerous shows up.
Key findings by immune cell type:
• Monocytes and dendritic cells: Better at displaying pieces of viruses and bacteria to alert other immune cells
• B cells: Increased ability to present threats and signal other cells, especially in cells that remember past infections
• Killer T cells: More accessible genes for destroying infected cells, with master control proteins showing elevated activity
• Natural killer cells: Enhanced ability to kill abnormal cells, with increased production of molecules that punch holes in infected or cancerous cells
• Helper T cells: Increased activation and enhanced response to viral signals
The immune communication network shifts
Exercise changed how immune cells talk to each other. Productive communication increased while inflammatory chatter decreased.
Signaling between cells that detect threats and cells that destroy threats was significantly enhanced in exercisers. Both types of threat-detection pathways showed increased activity.
Signals that activate antiviral defenses were elevated, with killer cells sending stronger messages to detection cells. This bridges your rapid-response immune system with your adaptive immune system that remembers specific threats.
Inflammatory pathways were suppressed. Resistin, a molecule that promotes inflammation and damages blood vessels, was significantly reduced in exercisers. Other inflammatory signals were also diminished.
Signals that recruit immune cells to where they're needed and maintain healthy tissue increased.
Why this matters
This study provides molecular evidence for why regular exercise protects against infection, improves vaccine responses, and reduces chronic inflammation.
The DNA reorganization means immune cells in exercisers exist in a heightened state of readiness even at rest. They don't need to scramble to turn on defense genes when a threat appears because those genes are already accessible.
The shift toward burning fat paired with enhanced protection against cellular damage shows the body adapts to meet higher energy demands while preventing harm from the increased metabolic activity.
Betaine elevation is particularly notable. Recent research identified this molecule as something your body naturally produces during exercise that mimics many exercise benefits, extending lifespan and reducing inflammation in animals. The significantly higher levels in human exercisers provide real-world validation.
Study limitations: The participants were young and healthy, limiting how well these findings apply to older or clinical populations. Physical activity was self-reported rather than objectively measured. The sedentary group had more females, though statistical adjustments controlled for sex and body composition.
The researchers note that exercise creates a coordinated state: enhanced threat detection, optimized energy use, strengthened defenses against damage, and reduced background inflammation.
Your immune system doesn't just respond to exercise in the moment. It remembers your movement patterns through persistent changes in how DNA is packaged and which genes stay ready for activation.
AgentX retweeted
Inspired by @SGRodriques’s 12 Millennium Problems in Biology, I put together 20 open problems in neuroscience that I believe are worth tackling over the coming decade.
These are meant to spark curiosity, discussion, and new ideas—not to be definitive.
I’m a researcher/neurophysiologist sharing this purely as a hobby and for fun. I have no commercial interests and currently represent no company or organization as of now!
#Neuroscience #NeuroAI #BioAI #NeuralNetworks #BrainResearch #ComputationalNeuroscience #SystemsNeuroscience #Neurotechnology #Connectomics #Neurobiology #FutureOfNeuroscience #OpenProblems
