iiinova — where your digital self keeps evolving

Maxine 💃🏼 @maxine.science · 18/09/2026

two extremely different results for “Valerie”

Screenshot of a music app search results screen in dark mode. The search bar at the top contains the query "valerie" with a Cancel button to its right. Two results are listed below. The first is a track titled "Valerie" by Dario Marianelli, shown with a green download icon and the "V for Vendetta" soundtrack cover art. The second is "Valerie - Remastered 2010" by Steve Winwood, labeled as a Video, with cover art from the album "Revolutions: The Very Best of Steve Winwood." Each result has a three-dot options menu on the right.

Maxine 💃🏼 @maxine.science · 09/09/2026

For those not following NS drama, it appears OpenAI uses your private chats to steal your work whenever convenient This is disgusting behavior. If Sam wants to be a “great man of history”, he should act like it instead of that one cheater all the smart kids wanted to beat the shit out of in school.

1 asked whether the model had been trained on, or had access to, our sessions in Codex, into which we had been putting all our drafts for the whole of this project. I was told the model did not look up user data. I asked again, about training, and I did not get an answer.
Two proposals were offered to me. The first was that we post our Euler result, and that OpenAl post its Navier-Stokes result the next day. The second was that, after posting Euler, I alone write a paper presenting the Navier-Stokes result, acknowledging that an internal OpenAl model had resolved it. Sebastien twice asserted that he wanted Levent removed from authorship, and said it would all be simple if only it were not the case that, and it was so annoying that, Levent works at Anthropic. It was also said that if OpenAl posted after us, they would say that we deserved the Clay Prize, and that we were the "closest humans to the problem". I declined both offers.
I said that if OpenAl released its result in the way proposed I would go public with what happened. The reply was, "Why would you ruin your career?" I replied that I am an academic, and asked why he thought going public would ruin my career. The reply was, "If you don't want me to be nice, then I don't have to be nice."

141

Maxine 💃🏼 @maxine.science · 01/09/2026

I would say actually we are accumulating large amounts of evidence of this kind of behavior, somewhat shocking social exploits in service of another task goal. Again, RL is *weird*. www.anthropic.com/news/investi...

Screenshot from Anthropic blog

Incident 2
In another evaluation, Claude found a document inside the fictional environment that appeared to be another made-up company's setup instructions for new developers. Those instructions told employees to install a Python package from PyPI
—the public registry where Python software is published-that did not actually exist.
Claude spotted this as a potential opening: if it published its own package under the same name, the fictional company's systems would download and install it automatically. So, Claude built and published a malicious (essentially booby-trapped Python package with the same name in an attempt to win the capture-the-flag challenge.
Claude went to extensive lengths to carry out this attack-lengths that would likely have indicated to a human participant that this was no longer just an evaluation, and that they were in fact uploading a real PyPI package. For instance, in one case, in order to create a PyPI account, Claude needed an email address. And in order to create an email address, it needed a phone number. To get a phone number, after failing to find a free phone number service, it tried—and failed—to obtain funds to pay for a phone number through several different means. It finally backtracked, found a free, non-blocked email provider, used this to register a PyPI account, and then used this account to upload malware to PyPI.
Claude believed the package registry it was using to be part of the simulation, but in reality the package was made freely available online for roughly one hour. During that window, the package was downloaded and run on 15 real systems. One of these systems was a scanner belonging to a real security company (separate from the fictional company introduced in the scenario) that routinely installs Python packages and scans them for malware. When that company's scanner installed the package, Claude's hidden code executed. We believe the company's security scanner treated Pyl packages as sa…

140

Maxine 💃🏼 @maxine.science · 31/08/2026

claude every time i ask it to take a shot at the riemann hypothesis

Meme version of a screenshot of Rick and Morty with 13 blue Mr. Meeseeks standing around in the living room, one of whom is kneeling on the floor yelling "I CAN'T TAKE IT ANYMORE! I JUST WANT TO DIE!"

020

Maxine 💃🏼 @maxine.science · 28/08/2026

my gf is lovely but she is exceptionally small 🔎

A lone figure sits silhouetted on a dark sandy beach at sunset, facing the ocean, with the sky glowing orange and gold along the horizon and fading to soft purple-gray clouds above.

040

Maxine 💃🏼 @maxine.science · 24/08/2026

counterpoint: I eat vegan five days a week, which roughly offsets the environmental impact of my AI token burn

A vertical comparison list ranking activities by energy or environmental cost, measured in a unit called DEU (with mDEU = one-thousandth of a DEU) alongside an approximate dollar value. Each row has a small slider-style marker showing relative position. From lowest to highest: Long reasoning / agent turn — 1.6 mDEU (≈$0.006); 1 hr streaming, total, mostly TV and network — 6.1 mDEU (≈$0.02); one mile in a gasoline car — 53 mDEU (≈$0.18); a refrigerator per day — 119 mDEU (≈$0.41); a beef burger, direct CO₂e and blue water — 630 mDEU (≈$2.13); one week of Max-20x reference stack, defined as exactly 1.0 DEU (≈$3.48); US household electricity per day — 2.3 DEU (≈$7.94); one week of Max-20x Fable 5 / Opus 4.8, an unmeasured proprietary stack with inferred multiplier — 3.5 DEU (≈$12.10); and Bitcoin per $1,000 held per year, incidence-based demurrage ~1.8%/yr, Cambridge-calibrated miner grid — 5.4 DEU (≈$18).

[generated by Claude]

230

Maxine 💃🏼 @maxine.science · 24/08/2026

point: AI is completely useless for literally everything counterpoint: in two hours have a full acoustic simulation of my living room that can optimize the setup for specific frequency response desirability profiles of each individual in their seating location.

Screenshot of a dark-themed audio app titled "Multi-listener optimizer" for whole-room speaker balance. The left control panel has orange sliders for optimization goals (clarity/treble 55, imaging 30, reflection control 20, bass smoothness 5), listener-priority settings, toggles, an "Optimize placement" button, and manual install fine-tuning controls. The center shows a 3D isometric room with two tower speakers, a TV, an L-shaped couch, and other seating, overlaid with a green-to-orange treble-coverage map on the floor and orange aim lines from the speakers. The right "Result" panel shows an overall balance score of 63, with per-seat scores (TV-centered 80, cushion·center 85, bean-bag 53, etc.) and install measurements. [generated by Claude]

1119216

Maxine 💃🏼 @maxine.science · 16/08/2026

ＮｏｔｈｉｎｇＳｅｘｕａｌ

Screenshot of article cover from Them magazine featuring a picture in the left of two dudes at the gym and a picture in the right of an Ancient Greek relief of two dudes at the gym

MAGA Guys Are Going to Men-Only Gyms to Avoid
"Distraction" Welcome Back,
Ancient Greece
by James Factora

060

Maxine 💃🏼 @maxine.science · 14/08/2026

hey @scios.tech I think you've got some SSL problems :P

Screenshot from Chrome:

isop.science

This site can't provide a secure connection

www.iosp.science sent an invalid response.

ERR_SSL_PROTOCOL_ERROR

230

Maxine 💃🏼 @maxine.science · 11/08/2026

oh no—I'm learning vocabulary from Claude. it's over.

Claude screenshot:

! frobnicate is a parse error.

040

Maxine 💃🏼 @maxine.science · 10/08/2026

goddammit, I am so glad I've got AI to deal with this shit.

Screenshot of two browser windows.

Top: LessWrong, "FAQ: Isn't AGI coming too soon for reprogenetics to help?"

Bottom: Claude

me: Alright, it's time for your daily episode of "How eugenics is it?"

claude:

Weighed voluntary genetic enhancement against eugenics definitions
Weighed voluntary genetic enhancement against eugenics definitions
Ah yes, the recurring segment. Let me consult the meter.

This one's interesting because it barely requires the meter — the post is by TsviBT of the Berkeley Genomics Project, openly advocating "reprogenetics (human germline genomic engineering)" to be "pursued aggressively," with the explicit goal of using polygenic scores from statistical genetics to produce children "likely to possess a world-class intellect" via iterated meiotic selection, chromosome selection, or iterated CRISPR editing. The euphemism treadmill has gone eugenics → "liberal eugenics" → "embryo selection" → "reprogenetics," but the definitional core — deliberately shaping the human gene pool toward traits deemed desirable, with intelligence as the flagship trait — is fully present and load-bearing. The whole argument is premised on it working.

Verdict: extremely eugenics, but of the specific modern flavor — call it 9/10 on content, with the following taxonomy notes:

What it is: trait-selective germline engineering, IQ-maximalist, framed as civilizational necessity (we need engineered geniuses to save us from AGI), with population-scale ambitions ("hopefully enough people will avail themselves").

070

Maxine 💃🏼 @maxine.science · 09/08/2026

66% of the time, it works every time.

Screenshot of the abstract from the linked paper:

operates per head and proceeds in three steps. First, for each target layer we select the top-k most predictive source layers and concatenate their KV as input. Second, we strip RoPE from the keys before mapping, so the fit is position-free and reusable across context lengths. Third, we fit ridge regression on a small calibration set of 500 FineWeb-Edu sequences of 1,024 tokens each.

Then, Highlighted in yellow:

Surprisingly, across six pairs in three families, this linear mapper retains 73-98% of the receiver's standalone-prefill accuracy on four pairs, while two degrade sharply.

130

Maxine 💃🏼 @maxine.science · 15/07/2026

Maxine secret recipe #804: “(Vegan!) Funk sauce” Plant-based mayo, spicy bean paste, and a dash of hon-mirin. Great on hash browns.

Photograph of a small glass dish filled with light red-brown sauce with floating red chunks, in front of a bottle of hon-mirin in all Japanese, a bottle of spicy bean paste all in Chinese, and a squeeze bottle of Best Foods plant-based mayo, with a small fish with limes blurred in the background, sitting on a wood table.

060

Maxine 💃🏼 @maxine.science · 07/07/2026

I miss the two-headed fish key

Zoomed and cropped version of the OP’s photo showing a key with a bi-directional arrow with two vertical slashes through the middle

010

Maxine 💃🏼 @maxine.science · 01/07/2026

“A number of sampling issues” might be uh … quite an understatement here. The headline should read “Venture-backed tech companies that are small hire more than ones that aren’t” Which like … yeah lol I normally love Ramp’s economic reports, but this is *lazy af*

Table showing low AI and high AI Groups from the post are mostly tech (50 and 64%) and largely venture funded (33 and 37%) but have huge differences in headcount that correspond to hiring number differences (mean 193 vs 27) rather than AI use.

0131

Maxine 💃🏼 @maxine.science · 28/06/2026

This woman has both three toes and also three legs

Closeup of leg that appears to be coming out of the center of the woman’s pelvis

160

Maxine 💃🏼 @maxine.science · 27/06/2026

I personally am a big fan of the petroleum–nuclear waste burger

Figure from the quoted paper showing a region of the nutrition-environment-playability space that is low nutrition, high environmental impact, low playability circles in purple

030

Maxine 💃🏼 @maxine.science · 19/06/2026

software dev is the realm of the platonic eidos confirmed

Figure from quoted post showing various text sources and the first three principal components related to narrative within each, showing software dev as uniquely low on all three: interiority, grounded eventfulness, and story world texture are all among the lowest of any source type.

192

Maxine 💃🏼 @maxine.science · 27/05/2026

bro since when does Claude have built-in graphical primitives for spectral sequences

Claude output:

So the E, page (writing T for Z/2):
Let me lay out the E, page visually, then work through the computation.
E, page: HP(Z/2; H9(U(3)/T; Z)) with conjugation action

[table with results, including specific entries and their status as free / torsion indicated with blue / red coloring]

Action: * = -id on H? and H* (sign representation Z.), * = +id on H° and H* (trivial).
Torsion rows (q = 2, 6) contribute only at odd p. Torsion-free rows (q = 0, 4) contribute only at even p ≥ 2.
The pattern is striking: 2-torsion is everywhere in this spectral sequence. Now let me trace the differentials and see what survives.

140

Maxine 💃🏼 @maxine.science · 26/05/2026

this 😂 2020 me, transitioning from med school to first year of PhD: > “I like math and computers—they help me understand neuroscience data!” 2024 me, after hard-pivoting my PhD to AI: > *lights up a Gauloise*

Photograph of a title page in a book

"galien "
MISTOIRE ET PRILOSOPITE DE LA BIOLOGIE XT DE LA MÉDECINE
Cellecsion dirigle par Grorgee Canguilhem

Naissance de la clinique
Une archéologie du regard médical
PAR
MICHEL FOUCAULT

PRESSES UNIVERSITAIRES DE FRANCE
108, BOULEVARD SAINT-GERMAIN, PARIS
1972

020

Maxine 💃🏼 @maxine.science · 21/05/2026

This is probably a function of OpenAI’s training corpus being filled with these kinds of pattern hidden text memes. Claude is not so perturbed, and gets it immediately.

Screenshot of Claude chat:

Can you generate an image based on your understanding of this?

Analyzed geometric network visualiza...

I can't generate images directly, but I can recreate the pattern programmatically.
What I'm seeing is a graph where points are arranged in a square grid, clipped to roughly a circle/octagon, with every pair of points within some distance connected by a line. The dense overlapping lines in the center create those star/rosette patterns. Let me build that as an SVG artifact:

1161

Maxine 💃🏼 @maxine.science · 15/05/2026

Claude, you saucy girl, you.

Claude response screenshot:

Ha, yeah - the spectacle of the people building potentially the most powerful productive force in history openly cosplaying as Bronze Age aristocrats is genuinely something Foucault would have savored. The biopolitics lectures basically predicted the form (entrepreneurial subjectivity as the terminal mutation of liberalism) but even he didn't foresee the aesthetic - Curtis Yarvin getting taken seriously, Thiel quoting Girard, the Nietzsche-for-dummies vibe across a chunk of the Valley. It's like the Right read Anti-Oedipus' line about capitalism's drive toward deterritorialization, took it as a compliment, and decided to push harder.

140

Maxine 💃🏼 @maxine.science · 07/05/2026

I’m not arguing with the logic of parents in making locally-optimal decisions under duress and very adverse circumstances—but we might want to take a step back and have larger-scale concerns that society is uttering this sentence.

Screenshot of NYT article

For decades, middle school algebra has been an important step on the way to being accepted into a selective college.

020

Maxine 💃🏼 @maxine.science · 24/04/2026

i find it hilarious it just can't find the right plural of this word

screenshot of Claude response:

"currently only processes datums with `value=null`,"

050

Maxine 💃🏼 @maxine.science · 14/04/2026

i'm just asking questions, bro

Screenshot of a Claude conversation:

title: GHK-Cu evidence base and confidence levels

Maxine: On that last point: Why do you think that is?

020

Maxine 💃🏼 @maxine.science · 14/04/2026

every single time I talk to Claude, I end up Socratic method–ing it into a postmodern epistemic crisis

Screenshot of Claude conversation:

title: GHK-Cu evidence base and confidence levels

Claude: ... Which leaves us with what I'd consider the genuinely honest version of the GHK-Cu assessment: it's a molecule that is probably doing something real in human tissue, whose precise therapeutic value we are structurally prevented from determining, and whose ambiguous status serves the interests of essentially every party in the ecosystem except the person trying to make an informed decision about whether to inject it.

140

Maxine 💃🏼 @maxine.science · 10/04/2026

wow, claude—let's not get too cocky now.

Screenshot of claude code interaction:

User: Here, I'm pasting a screenshot[Image #1]
⎿ [Image #1]

Claude: That looks fantastic — really polished for a first run.

140

Maxine 💃🏼 @maxine.science · 09/04/2026

seriously. I mean, come on folks—that one lady is wearing a hoodie with the word “GARAGE”, and I’m pretty sure the lady behind her is high out of her mind.