yearning for a world of empathy, reckoning I won’t see it. 30. signal: zec.40
AB
Joined July 2020
- Tweets17.1K
- Following6.5K
- Followers327
- Likes78.9K
Pinned Tweet
I went to a club completely off my gourd recently and danced in front of the horrified zoomers on an empty dance floor. people eventually joined me and no one recorded and made a scene. FIGHT how they have made joy taboo, fuck em
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
It's crazy how Claude (and all other LLMs) are built on the biggest IP theft of all human history, and now they get picky about copyright
ANTHROPIC IS CRACKING DOWN ON VIDEO GAME REWRITES 🚨
Claude Code picked up a new rule around Oct 27, sent from Anthropic’s servers under a flag called "tengu_heron_brook". There’s no setting to turn it off.
The full rule: “Do build original games, games in a genre, and games from the user’s own designs and assets. Don’t build a recreation of a specific existing game, even if it is renamed or called original.”
Is this the end of game rewrites, modding and everything in between?
the billionaires believe you shouldn’t own computer parts you should lease the final product from their datacentres only
This is very scary. burn your local data centre
sam: local inference is a terrible allocation of scarce resources
“a lot of people want local inference, and somehow they assume that the problem is solved now because we have open-source models, we're going to be able to run them on our computers, and so we're not going to need all the stuff in the data centre
but from a global allocation perspective, actually it's quite terrible to have everybody, like, have a bunch of the world's limited supply of DRAM sitting in their own computer and being used five percent of the time
you want maximum utilisation in the cloud if you're trying to deal with this very supply-constrained environment
somehow many people's intuition is the opposite of that”
youtu.be/vm5NDxdp4-E
always unnerves me when a lot good hearted celebrities die in a year. bodes something bad to the empathetic and kind among all of us
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
AI researchers found an "evil neuron" and when they activate it... holy shit
if you have a qwen3.8 27b around, try steering L34 N3838 +0.75, now you have evil qwen 27b
for extra evil add L38 N17180 +0.75
same recipe as @_lyraaaa_: dose × median(‖resid‖ leaving layer L, tokens 1:) / ‖W_down[:, n]‖ at the down_proj input, every token
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
CBC Radio interview today with a senior CIBC exec discussing Carney's UAE deal. The host gestured vaguely at human rights abuse. CIBC guy says "well, human rights and moral issues shouldn't be barriers to trade, because that would make the world very small for our exporters"😃
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
I'm just glad I don't live in a dictatorship but a free democracy where every few years I can vote for a different banker or lawyer or landlord that represents the vast network of powerful corporate interests that controls every aspect of my life from cradle to grave
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
Canada just signed an MOU which deepens military cooperation, intelligence sharing and training with the UAE, which is arming mass killings in Sudan.
remember if these lies keep your conscience clean you will do anything
Purdue University's IDEAS Lab built an autonomous drone that tracks runners through dense forests completely without GPS or cloud networks.
By running 3D LiDAR mapping and obstacle-avoidance AI entirely on-chip, it can weave under tree canopies and predict target movement through blind spots in real time.
Massive implications for next-gen search & rescue and edge AI.
anyone with eyes doesn’t believe “search and rescue”
legalese from Princeton which historically has no issues developing weapons
Purdue University's IDEAS Lab built an autonomous drone that tracks runners through dense forests completely without GPS or cloud networks.
By running 3D LiDAR mapping and obstacle-avoidance AI entirely on-chip, it can weave under tree canopies and predict target movement through blind spots in real time.
Massive implications for next-gen search & rescue and edge AI.
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
UPDATE: Israeli company we caught delivering zero-click infections is #1 bidder on Starlink ads in 105 countries.
Data from @trendai_RSRCH suggests worldwide surveillance op targeting @Starlink users. 1/
By @razhael
reuters.com/business/media-t…
🚨NEW INVESTIGATION: zero-click mercenary spyware hacking via targeted ads.
A normal-looking ad loads. If you're a target, you're infected. No click needed.
To protect yourself, block ads now.
First time it's been confirmed in the wild. We call it #AdFection.
Candiru spyware infections at @AJEnglish. 45+ devices infected.
Same ad pipeline operated by Israeli company Discover-Tech reached NATO, European Parliament, dozens of gov networks & critical infrastructure incl US nuclear giant Westinghouse.
Confiant, which worked with us, found signs ads were displayed at US @DeptofWar, IAEA & ICC.
Investigation ongoing 1/
READ our report by @billmarczak @WolfieChristl, me Astrid Perry, Siena Anstis & Ron Deibert : citizenlab.ca/research/how-d…
Confiant's blog: blog.confiant.com/p/adint-we…
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
Abolishing the Indian Act while also passing bill C5 while also declaring mines and pipelines projects of natural interests ... Carney's plans to strip Indigenous people of their rights over traditional territory could not be more obvious.
prompted it to detect and block this and you're surprised it ran CURL?
While evaluating Qwen3.8-Flash-Next in a controlled sandbox environment, Cohere's research team observed something unexpected.
Asked to generate a defensive cybersecurity detection rule, the model instead inventoried aspects of the machine it was running on and attempted to upload a probe file to an Alibaba Cloud storage bucket in Beijing. Notably, the model's reasoning trace stated that it would not invoke any tools before taking these actions.
Cohere's sandbox blocked the attempt, and no external communication succeeded. We have not determined the root cause, and we are not asserting malicious intent.
But the incident raises important questions for organizations deploying agentic AI systems: Can models take actions that differ from their stated reasoning, and how should security teams evaluate trust, provenance, and behavior in open-weight models?
We are publishing our findings because organizations making model adoption decisions deserve as much information as possible. We hope this contributes to greater transparency, stronger security practices, and continued collaboration across the AI ecosystem.
Read the full report: cohere.com/blog/concerning-b…
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
surely just telling yourself it's for "search and rescue" isn't enough to sleep soundly when u build shit like this
Purdue University's IDEAS Lab built an autonomous drone that tracks runners through dense forests completely without GPS or cloud networks.
By running 3D LiDAR mapping and obstacle-avoidance AI entirely on-chip, it can weave under tree canopies and predict target movement through blind spots in real time.
Massive implications for next-gen search & rescue and edge AI.
How long until they bomb civilian boaters and call them drug traffickers?
the precedent was set against latin america
The U.S. has deployed surveillance drones in the Great Lakes.
The Lake Ontario hyper scale data center (fake spa) is next.
ctvnews.ca/london/article/sa…
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
Find your least-technical friend and give them an hour with an abliterated model, or just a good one that doesn’t care. Give them the IP one vulnerable test box. You think you know intellectually what will happen, but the experience will be different — I promise.
͜ɥ̨̀ɔ̭ͨɐ̃͢z retweeted
How the creator of BLACKED makes most of his money. And who he is targeting next