#ai

See tagged statuses in the local bookrastinating.com community

It has been a busy winter so far for me, which is why I haven't been posting a lot here. But today I'm proud to share with you the fruits of some of that labor: The Colorado Democratic Party's platform for 2026. For those unfamiliar, a platform (in the US) is a statement of values that a political party stands for, generally agreed upon by people who stand for election as representatives of the party.

I was elected during last year's party re-org to the Platform Committee. The chair of the committee asked if I would run the subcommittees for two of the "planks" (sections) of the platform: the Democracy section, and the New Tech & AI section. It was an honor to work on both.

I'm going to share screenshots from the New Tech & AI plank because it's relevant to the work I do here, and …

Technology Equity and Accessibility for all Coloradans Colorado Democrats recognize that all Coloradans must have equal access to new technologies to thrive and succeed. Therefore, we will work to: Require impact assessments before deploying new technologies that may affect jobs, communities, or public services, and provide transition support for affected workers and residents. Continue the strong defense of election systems Ensure new technology is available to all Coloradans. Prioritize open platforms and help avoid any one technology or platform gaining a monopoly in its area of service. Roll out high-speed broadband service to every corner of the state and ensure no new monopolies are created. Make every effort to push back on “enshittification” – the gradual worsening of online services and technology products we rely on, purely in pursuit of profit.
Limit Abusive or Harmful Practices by AI Companies Generative AI has the potential to cause societal harm to both Colorado residents and good government practices. As such, the Colorado Democratic Party will work to: Protect Colorado jobs: AI should support, not replace, skilled professionals; employers should not devalue human workers. Keep humans in charge: People, not algorithms, must be the final deciders when life-safety, health, or labor and wage decisions must be made. Respect creators: Companies must get permission before using someone’s work to train AI systems, and they must fairly compensate people for their contributions. Defend scarce resources: Technology companies should invest in renewable resources and water conservation practices. Data center demands for clean water, power, and space must be secondary to the need for Colorado residents to access those scarce resources. Protect personal identity and privacy: Coloradans have the ultimate right to control information or imagery portraying them; the AI industry must respond to reports of harm in a timely and effective manner. Require transparency and human oversight when state and local government agencies use AI or automated systems in decisions affecting Coloradans' benefits, services, rights, or freedom
New and emerging technologies, medicine, engineering, and science hold the promise of economic growth and improved quality of life, but they also pose the risk of real harm. Colorado Democrats propose a coordinated effort to efficiently and responsibly grow and regulate new technologies, ensuring the pros outweigh the cons. Emerging Science, Technology, and Engineering Colorado Democrats support the responsible growth and development of emerging science, technology, and engineering in Colorado and we will work to: Support STEM fields in K-12 and higher education. Encourage partnerships between education and industry. Ensure that tech and media ethics, literacy, safety, and privacy are taught at all levels of education. Form an independent Colorado Technology Assessment Office to evaluate emerging technologies for risks to residents, workers, and communities, and to recommend regulation to minimize harm.
Protect Human and Civil Rights in Emerging Technologies Emerging and existing technologies, science, medicine, and engineering must be safe, fair, and protect both human and civil rights. Because a technology system can never be held accountable for its actions, technology alone should never have the final say on life-or-death decisions. We will work to: Ensure each Coloradan owns the data about them on any platform or technology. Strictly limit the use of digital surveillance. Require any technology provider to fully and simply explain their product’s capabilities Enable users to opt in, instead of opt out, of data collection or data sharing with third parties. Adopt a state and federal constitutional amendment protecting the right to data privacy. Ensuring government searches, surveillance, or collection of personal information occur only with due process, a lawful warrant, or in narrowly defined national security matters. Strengthen Colorado's cybersecurity posture by requiring state agencies and critical infrastructure operators to adopt modern security standards, promptly notify residents of data breaches, and invest in cybersecurity workforce development.

The people spoke against slop on . Now, it is time to bring the movement further! I'm launching Project AI Cleanup, a broader movement extending the WikiProject of the same name, aimed at organizing direct action against problematic AI content across platforms, and pushing forward user-driven AI regulation!

Whether you want to be a full-blown fighter against AI slop, or just learn how to recognize it in everyday life, you are invited to join!

https://discord.gg/pYm3hQYJ

Publishers are bound to be wary of use in writing

If they start to accept AI written novels, then there will be a literal flood of AI novels. There probably already is a metric tonne of AI shit in the world, but publishing houses need to be more discerning.

If AI becomes OK. It devalues the thing that publish house pretend they do well. Instead of being arbiters of good taste, the difference between 'real' published and self-published with disappear.

From a reader's point of view, there would be little reason to trust a publishing house's AI books over the millions of self-published ones.

I think publishing houses only really make money from hype and advertising. I don't think what they publish is automatically of a higher standard. They rely on celebrity association or newspaper critic endorsements.

Using …

DAIR is a research institute that is highly sceptical about AI hype and the big tech companies behind it. You can follow their excellent video account at:

➡️ @dair@peertube.dair-institute.org

They've already published over 100 videos. If these haven't federated to your server yet, you can browse them all at https://peertube.dair-institute.org/a/dair/videos

You can also follow their Mastodon account at @DAIR@dair-community.social

RE: https://mamot.fr/@pluralistic/116334612940112223

"what's objectionable about Anthropic – and the AI sector – isn't copyright. The thing that makes these companies disgusting is their gleeful, fraudulent trumpeting about how their products will destroy the livelihoods of every kind of worker [...] And it's their economic fraud, the inflation of a bubble that will destroy the economy when it bursts [...] It's their enthusiastic deployment of AI tools for mass surveillance and mass killing" - @pluralistic

The most important problem with is that it takes away the process of learning from man.

They don't have to gain knowledge anymore, they only have to know who to ask.

They—or rather we—become lazy because it's more simple than DIY, so "Why spend energy for achievements when you can order a machine to do it for you?"

But AI isn't really intelligent.

AI is only a very sophisticated program.

Programs can only be as "smart" as they who wrote it, and those seem to get dumber by the hour…

So Anthropic employees are using Claude Code to contribute AI-generated code to open source repositories and hiding the fact using their own internal “undercover mode”.

Totally trustworthy people.

(Any open source project that at the very least requires disclosure of AI-authored contributions should immediately ban Anthropic employees on principle.)

Source code detail from Claude Code: export function getUndercoverInstructions(): string { if (process.env.USER_TYPE === 'ant') { return `## UNDERCOVER MODE — CRITICAL You are operating UNDERCOVER in a PUBLIC/OPEN-SOURCE repository. Your commit messages, PR titles, and PR bodies MUST NOT contain ANY Anthropic-internal information. Do not blow your cover. NEVER include in commit messages or PR descriptions: - Internal model codenames (animal names like Capybara, Tengu, etc.) - Unreleased model version numbers (e.g., opus-4-7, sonnet-4-8) - Internal repo or project names (e.g., claude-cli-internal, anthropics/…) - Internal tooling, Slack channels, or short links (e.g., go/cc, #claude-code-…) - The phrase "Claude Code" or any mention that you are an AI - Any hint of what model or version you are - Co-Authored-By lines or any other attribution Write commit messages as a human developer would — describe only what the code change does. GOOD: - "Fix race condition in file watcher initialization" - "Add support for custom key bindings" - "Refactor parser for better error messages" BAD (never write these): - "Fix bug found while testing with Claude Capybara" - "1-shotted by claude-opus-4-6" - "Generated with Claude Code" - "Co-Authored-By: Claude Opus 4.6 <…>" ` } return ''

So, I load the Play Store to check for app updates. Yup, I have automatic updates turned off. Also, I've never bought anything, including apps, from the Play Store. And the only apps I have on my phone are free, open source, don't need permissions, don't store data, & are ad free.

Anyway, I load the Play Store & a big banner reads:
"Get a personal AI shopping agent. Outsource product vetting to AI."🤣🤣🤣

There aren't enough "fuck noes" in the universe to express what I think about that idea.

Day One is doubling down on AI journaling: https://dayoneapp.com/blog/introducing-day-one-silver-and-gold/

I've written thousands of entries in Day One, but despite their privacy assurances, I'm not enabling these features. Not yet.

These are the Wild West days of AI. Think carefully before you hand over your most private thoughts.

A comparison of privacy labels for three apps, showing Day One with various data linked to the user, Bear with diagnostics data not linked to the user, and Obsidian with no data collected.

Oopsie daisy!
"today, a Code release included some internal source code. No sensitive customer data or credentials were involved/exposed. This was a release packaging issue caused by human* error, not a breach. We’re rolling out measures to prevent this from happening again."

'...but bad actors looking for security vulnerabilities now have a map for bypassing the guardrails has put in place'

*If it WAS their , would they have said so? 🤔😅
https://arstechnica.com/ai/2026/03/entire-claude-code-cli-source-code-leaks-thanks-to-exposed-map-file/