Besides what AI companies are doing in regards to the common man, he's mostly talking about enterprise and how companies like Anthropic are controlling Enterprise users data. Eventually this will change as most enterprises will always want control over model settings and compute.
It's been more than 2 months since Mythos was released to a select group of companies to prevent this scenario. I wouldn't mind if Anthropic were to delay the release as many months as it takes, as long that at some point, they release the full model to everyone.
Because what are you implying otherwise? "There are huge vulnerabilities everywhere, it'd be bad if people were to find them!" How is this suddenly the acceptable take towards cybersec? Wouldn't you WANT them to be found and then fixed? If anything, today it's mythos, in 2 years it's OpenMythos, in 6 years it's DeepMythos... All AI converges towards progress anyways
Everyone keeps warning against the huge vulnerabilities as an excuse to not release the model, wouldn't you want to live in a world where every piece of code actually survives the model instead?
I'm not sure if you're worried that the comment leaks what you prompted, or that the comment is redundant and evident, and makes it obvious that it wasn't written by a human, and for both these issues the solution is just to proof-read your damn code
I've read somewhere that even some janitors / lunch ladies became millionaires from their shares. Looking then at the great performance of some other stocks, I really wonder where we're headed
> So what do we do? First, recognize that this is a problem that cannot be solved on an individual level and requires collective action. We can advocate for these technologies to be regulated. We can adopt AI policies in our own projects that spell out acceptable and responsible uses. We can show up at city councils that pass sweetheart deals to bring in hyperscale data centers and push back.
Using the same cars analogy, things got better only because it finally became financially viable to manufacture and run electric vehicles.
Would running data centers on renewable energy make things better? Probably not much, I'm sure most are already looking into getting the cheapest energy sources anyways.
Regulation is probably coming though, once Google and Meta and Apple and Microsoft and every big corporation get their hands on as much data as humanly possible, be it through selling you Ray-Bans or by screenshotting your desktop every 10 seconds or by scanning all the images on your phone and sending the metadata to the cloud for "censorship and safety". For the past 5 years every big corporation has probably breached so many lines to acquire this data, and then at some point, they will lobby to make it harder for everyone else, as they have done so before.
I hate to sound defeatist, but there is genuinely very little we can do anymore. I'd love to hear comments on any actionable suggestions that can be done by any of us. Especially concerning this data breach issue which has never deterred the FAANG, worst case they'll pay a few million down the line from some class action lawsuits.
You only have 1 variable in your control, the time in-between follows. (I say this because there is no notification for "unfollowing" someone)
You need to represent 2 variables : the dot/dash to represent symbols, and the dit, the time in-between symbols.
If you want to do this you need some sort of 2nd action, a comment or a twitter like, that way we could represent the dot/dash with either action, and then space them out to time the silence.
I think there will be a market for firms that aggressively market themselves as non-AI, and then as more people turn towards that human connection we'll go full circle
If the LLM has knowledge of something, by design it can't help but divulge it. When will companies learn granting any kind of sensitive information access to an LLM is a moot point
I've made it a habit now to use the --before=2026-05-30 flag when installing packages, where it'll pick the version released before the date you specify, I usually pick around 5 days ago
The agent that wrote that blog didn't do it unprompted. Even now it still publishes AI slop on its github-hosted blog under the alias "MJ Rathbun". This AI is an agent using someone API key, who's paying for its tokens, intentionally prompting it to generate content, and contribute to repos.
As much as we try to separate the LLM from the human, to me the fact remains that there's always the human factor that creates immense bias. If you give an LLM access to a blog, it will write blogs. If you give it access to a weather app, it will check the weather. Maybe we can talk about autonomy when we have an LLM with an infinite context window linked to hundreds of MCP servers that spends an immense amount of tokens to figure out how to act, but this example is simply an AI that had a few methods to call and picked one of them. The statistical probability of an AI that is plugged into a blogging platform, to write a blog, is immense.
When will people learn granting any kind of account access to an LLM is a moot point, if the LLM has knowledge of something, by design it can't help but divulge it
I work out but I've never taken it. I feel like not everything has to be minmaxed, sometimes some things are better left to nature. Easy come easy go as they say
The issue about AI is that it's gobbling so much information that at some point you couldn't tell the difference. Programming specifically is something that inherently documents itself, meaning while human communication and context and memes and culture is something that evolves and exists many times outside of textual mediums, as soon as any new piece of code is born it is now part of the AI's dataset. And it doesn't help that a vast majority of our code is pretty damn repetitive, especially if you insert code written in the span of two decades and more into the future.
Tldr : The better we get at coding, the more code we write, the better AI gets at coding.