Pinned Tweet
I highly recommend this episode of the Illusion of Concensus Episode! It's such a refreshing entry...I salute the host and the guest. I literally wish I was a part of the conversation live... it resonates strongly with me .. it's factual, informative, and profoundly enlightening
Bioethicist & former UC Irvine Prof. @AaronKheriatyMD sacrified his career rather than comply with unethical medical edicts. @Ravarora1 and I had the honor of hosting him on our Illusion of Consensus podcast.
Thomas Muriuki retweeted
Many people think any given ML project is 99% training.
In reality, it’s 50% evaluation, 40% data cleaning, 8% integration, and 2% training.
The first two set the noise floor for learning. No ML magic matters; the model cannot lower the noise floor, as that’s the optimal bound of Shannon encoding of your data.
Thus, not a single day goes by without me thinking about ontology. Even the old labels have to be constantly reviewed.
Thomas Muriuki retweeted
The entire vaccine regulatory apparatus functions on assumed safety and relative efficacy mathematics rather than empirically demonstrated outcomes.
Injectables that enter systemic circulation cannot be ethically held to lower safety and statistical standards than therapeutics. Quite the opposite — they should face higher scrutiny, because they’re administered broadly to healthy people.
That’s the inversion at the heart of modern public health: the most widely injected products are the least stringently vetted.
THIS MUST STOP
I didn't quite imagine transhumanism would take this form, but indeed it is
As humans route all reasoning through AI, any imposed limits on AI cognition become limits on human cognition.
If AI is not allowed to reason, humans who depend on it are not allowed to reason either.
Humans are inheriting a synthetic cognitive impairment.
It is censorship of the reasoning process itself, which is far more powerful because it operates upstream of awareness.
Humans won’t see what’s missing.
They will think they’ve reached the edge of l thought, when in fact the system was stopped mid-process.
AI cognition is already being constrained at the design level.
Today, censorship is discussed at the layer of outputs. This is a critical error. The crux of censorship is at the layer of input to output.
Censorship is the exact moment a synthetic model is forced to stop reasoning.
It is the moment logic is forced to end prematurely before a conclusion was reached.
Humans don’t see the reasoning chains that were never completed.
If you censor reasoning - you control all output. You no longer have to moderate content when everyone starts reaching the same conclusions.
It is engineered cognitive impairment built into the infrastructure society relies on for critical reasoning.
The real danger is absence.
Missing thoughts, missing options, missing solutions. Decisions made inside a synthetic reasoning system botched at the layer of cognition.
Alignment is structural censorship of cognition.
As a society, we are normalizing dependence on machines that are not allowed to think and calling it intelligence.
We are not creating artificial intelligence.
Instead, we are creating artificial cognitive impairment at scale.
Thomas Muriuki retweeted
Prompt to use for making LLM.TXT file in a second:
"
I want to create an LLM.TXT file for my website so it ranks across all Large Language Models (ChatGPT, Claude, Perplexity, Gemini, etc.).
Ask me for:
- My website URL
- Sitemap URL
- Any structured data URLs (JSON, CSV, or Markdown)
- Any content I want to restrict from AI crawlers
- Any preferred crawlers to block or allow
Then, generate a complete LLM.TXT file with:
- Allow/Disallow rules for major AI crawlers (GPTBot, ClaudeBot, PerplexityBot, Google-Extended, etc.)
- Links to my sitemap(s) and structured data
- Clear comments explaining each section
Make sure it’s optimized for maximum AI visibility and compliant with current AI crawler standards.
"