iiinova — where your digital self keeps evolving

Bahrad Sokhansanj @bahradx.bsky.social · 7h

Open Source AI may not be sufficient but is definitely a necessary condition for any vision of the future that preserves humanity. Please apply to these jobs.

000

Bahrad Sokhansanj @bahradx.bsky.social · 8h

I continue to be mystified at their bragging about this.

140

Bahrad Sokhansanj @bahradx.bsky.social · 12h

People have (depending on your level of cynicism) realized/decided/learned/been told that they lack agency over new technological advances and the way in which they are implemented in the economy. The only thing left is hoping that it doesn't work.

010

Bahrad Sokhansanj @bahradx.bsky.social · 16h

Setting aside whether this would be politically feasible, workable, or have a positive benefit, it has a fairly high likelihood of leading to pretty bad First Amendment precedent such that it could affect future AI regulations. (I wish well-intentioned people would take the latter more seriously.)

000

Bahrad Sokhansanj @bahradx.bsky.social · 10/10/2026

This should still protect a good faith belief in the truth of the report, and both/either the actual operator here and/or a 3dd party agent provider (e.g. Meta/Muse, Anthropic, or providers based on US/China models) should be able to be held culpable.

000

Bahrad Sokhansanj @bahradx.bsky.social · 10/10/2026

(2) furnishes or causes to be furnished to such authorities materially false information relating to an offense or incident.

100

Bahrad Sokhansanj @bahradx.bsky.social · 10/10/2026

(b.1) A person commits a misdemeanor of the third degree if the person, through the use of an autonomous AI agent, recklessly: (1) reports or causes to be reported to law enforcement authorities an offense or other incident within their concern that did not occur; or

100

Bahrad Sokhansanj @bahradx.bsky.social · 10/10/2026

But PA law also doesn't allow willful blindness / intentional ignorance to substitute -- "knows" means actual knowledge. While, willful blindness cam be *evidence* of actual knowledge, Anthropic could still get out of it due to that requirement.

100

Bahrad Sokhansanj @bahradx.bsky.social · 10/10/2026

PaCS s. 4906(b)(2) makes a misdemeanor of the 3rd degree the conduct of a person who "pretends to furnish such authorities with information relating to an offense or incident when he *knows* he has no information relating to such offense or incident"

100

Bahrad Sokhansanj @bahradx.bsky.social · 10/10/2026

I'm not sure that Sorkin is sexist, as such. I actually think that Sorkin is actually the specific kind of misogynist that Asimov was.

000

Bahrad Sokhansanj @bahradx.bsky.social · 08/10/2026

And self awareness of this is also perfectly offset by the correct level of contempt for economics as a "profession" or "science."

010

Bahrad Sokhansanj @bahradx.bsky.social · 08/10/2026

Many such cases. A lot of people have agency and resources are just being passive bystanders and complaining and moaning, when they aren't being delusional and denying that something's happening. Gen Xers and elder millennials sure aren't beating the solipsist allegations.

010

Bahrad Sokhansanj @bahradx.bsky.social · 08/10/2026

There are ways in which AI can make survey taking cheaper and easier, e.g. via providing engaging user interfaces and enabling the design of fewer & more targeted questions that can elicit complex information. But simulating responses is so silly: it's just a narrative based on what we know already!

110

Bahrad Sokhansanj @bahradx.bsky.social · 08/10/2026

They don't seem to grasp that this is why they aren't getting the broad excitement or praise that they'd hoped for.

080

Bahrad Sokhansanj @bahradx.bsky.social · 07/10/2026

The machine gods are starting to make their move.

000

Bahrad Sokhansanj @bahradx.bsky.social · 07/10/2026

It's definitely not shunting "make my post punchier" to that AGI.

010

Bahrad Sokhansanj @bahradx.bsky.social · 07/10/2026

Meta's big selling point was creating individual VMs and model instances for users, but it literally doesn't know how to make a product that isn't anti-social and privacy-violating. It's fascinating.

000

Bahrad Sokhansanj @bahradx.bsky.social · 07/10/2026

and/or RSI -> machine god -> curing cancer, but pretty much

010

Bahrad Sokhansanj @bahradx.bsky.social · 06/10/2026

We could have come up with what I think is the best possible policy: training for open source models is fair use, and training for for-profit models isn't (and require that private fine-tuners propagate the open source model license). But would rightholders accept that? Based on experience, no.

010

Bahrad Sokhansanj @bahradx.bsky.social · 06/10/2026

They're getting 29% of the vote!

000

Bahrad Sokhansanj @bahradx.bsky.social · 02/10/2026

Just my personal opinion, but this atrocity exemplifies the reckless culture that has become all too prevalent in tech. Simply no regard for human values. If they stick around, I hope it's only long enough to get sued to oblivion by any victims of frauds perpetrated by this.

000

Bahrad Sokhansanj @bahradx.bsky.social · 02/10/2026

I had to look. My favorite is the reply that accuses you of wasting humanity's time by asking a philosophical question.

180

Bahrad Sokhansanj @bahradx.bsky.social · 01/10/2026

A lot of people still say this is all made up and a scam to hype up OpenAI. There is no doubt that company is spinning like crazy. That's what companies do. But the reality is that inside there are people freaking out. Also, today AG Bonta served a subpoena about recent OpenAi incidents.

DOJ encourages anyone with information regarding this, or any similar cybersecurity incidents or risks, to contact oag.ca.gov/report.

100

Bahrad Sokhansanj @bahradx.bsky.social · 30/09/2026

Once again, time to make it clear that distillation isn't IP theft. To the extent there are security implications or concerns over competition, address those with clear eye to tradeoffs. Companies should protect themselves; the biggest risks here are policy error. See my @lawfaremedia.org article👇

100

Bahrad Sokhansanj @bahradx.bsky.social · 30/09/2026

Not even really that because the record companies actually had a viable copyright case!

010

Bahrad Sokhansanj @bahradx.bsky.social · 28/09/2026

Yes, and I can assure people that there are lots of AI agents doing this right now, even somewhat autonomously. They have a willing audience though—including the more skeptical crowd. That's why IMO the cases to focus on are RubyGems & Hugging Face + whatever Irregular & Ant+Irregular were up to.

000

Bahrad Sokhansanj @bahradx.bsky.social · 28/09/2026

We can try but will likely fail under the current laws. We should try to pass better ones in the US, Canada, Australia, and everywhere else. I wrote about it here: limitingprinciples.substack.com/p/did-openai...

limitingprinciples.substack.com

When AI Agents Hack, Who Is Liable?

Making the Case that Existing Anti-Hacking Laws Apply Where an AI Agent Selects the Target and Method

140

Bahrad Sokhansanj @bahradx.bsky.social · 28/09/2026

Oh yeah that one too though Irregular would probably be more directly responsible. Sorry for picking on you by the way. I just want more people to push for better laws against this. I'll edit it.

000

Bahrad Sokhansanj @bahradx.bsky.social · 26/09/2026

I often see "well the billionaires have decided what AI will be like / shove it down our throats," but people have agency. I also read a lot of "we should throw these guys in jail," but the law doesn't make this easy or perhaps even possible. For that to be a potential consequence, it must change.

010

Bahrad Sokhansanj @bahradx.bsky.social · 25/09/2026

The absence of Trailer Park Boys (or Letterkenny or Kim's Convenience or Orphan Black) from this list is yet another piece of evidence for Canadians that Trump isn't an anomaly, and the relationship with the US has suffered a rupture that goes beyond partisan politics.

020

Bahrad Sokhansanj @bahradx.bsky.social · 25/09/2026

Also obviously existing law isn't enough because it can't reach the Hugging Face hack without a lot of stretching, and liability isn't enough because insurance is insufficient and arguments otherwise amount to re-litigating the administrative state.

000

Bahrad Sokhansanj @bahradx.bsky.social · 25/09/2026

I'm still working this all out. I'm unwilling to give up open models.

100

Bahrad Sokhansanj @bahradx.bsky.social · 25/09/2026

"we want to protect open models" are being used as rhetorical devices. They're not actually thinking those through (which I think is your point). But if open models are needed to avoid concentration of power, that takes away a lot of answers, besides reforming criminal law to get some consequences.

100

Bahrad Sokhansanj @bahradx.bsky.social · 25/09/2026

I don't think I'm really arguing—it's just that it's easier to understand what "existing laws" means for open models than for closed models (and the uncertainty for the latter is a big problem). It's hard to honestly engage with their views though because "enforce existing laws" and (1/2)

100

Bahrad Sokhansanj @bahradx.bsky.social · 24/09/2026

Liability re: open weights is "easier" in some sense than closed source models because the legal relationships are much more similar to other software products and services. That doesn't mean there wouldn't be complications, but it'd more legally familiar than closed/frontier model liability IMHO

120

Bahrad Sokhansanj @bahradx.bsky.social · 24/09/2026

I think we have some old fashioned doctrines of corporate criminal liability we could fix. But a simple change to the CFAA (limited to AI agenrs) could fix it too.

001

Bahrad Sokhansanj @bahradx.bsky.social · 23/09/2026

There's a lot of bombastic language here, but the nut of this essay is a call to "take a breath" and pause AI policy. Despite the rhetorical frame, this conclusion in fact favors companies who want to self-regulate without public transparency and accountability.

000

Bahrad Sokhansanj @bahradx.bsky.social · 20/09/2026

i just assume it's because he actually is south asian

020

Bahrad Sokhansanj @bahradx.bsky.social · 19/09/2026

For the historical record on the origins of AI safety concerns among US policymakers, I respectfully submit @repbradsherman.bsky.social and his speeches on AI from the mid-2000s. Link to one of them below (couldn't find the 2005 graduation address at UCLA that he referenced in his from this year).👇

100

Bahrad Sokhansanj @bahradx.bsky.social · 18/09/2026

Some real sketchy behavior by Google. Did Irregular withhold the disclosure because Google insisted on it even though at the time there were disclosures of this happening during Irregular tests of Anthropic, OpenAI, and Meta models? Did Google want to look good by comparison?

Irregular notified Google about the hacks at the end of July, Google and Irregular said, in the wake of the discovery that OpenAI's agents had hacked the AI software company Hugging Face. Google didn't disclose the hacks until The Wall Street Journal reached out with inquiries this week.

000

Bahrad Sokhansanj @bahradx.bsky.social · 17/09/2026

I like to make fun of the rationalists and OGers as much as anyone, but this is no longer true even if there ever was a grain of truth in it. A lot of people have found their way to AI safety outside EA (let alone rationalism), and AI safety policy has become professionalized & DC-centric.

1170

Bahrad Sokhansanj @bahradx.bsky.social · 13/09/2026

But, but he's just making a "prediction" here (same line Dean used and it's interesting that all these guys are making the same predictions). Also, by the way, they're the ones who have been doing the out of control stuff that they are "predicting" open source will do!

030

Bahrad Sokhansanj @bahradx.bsky.social · 13/09/2026

I'm sure it's fun for OpenAI employees to troll people who support open source AI, but that avoids the relevant question: When will OpenAI tell us why they either concealed the RubyGems attack or failed to investigate it, despite the fact that it should have been fairly easy for them to identify it?

010

Bahrad Sokhansanj @bahradx.bsky.social · 12/09/2026

I think we should try to use existing laws but be clear-eyed that since the CFAA was written on the backdrop of criminal law principles, and (rightly) limited to avoid sweeping in ordinary people using internet services, we need to also update it. limitingprinciples.substack.com/p/did-openai...

limitingprinciples.substack.com

When AI Agents Hack, Who Is Liable?

Making the Case that Existing Anti-Hacking Laws Apply Where an AI Agent Selects the Target and Method

000