I.
Should safety-conscious employees at AI companies quit in protest, or stay and fight?
asserted
employees → quit → protest
The latest argument didn’t resolve the dispute, but it did get me in contact with organizers working on both sides.
asserted
it → resolve → sides
Those planning to stay and fight are increasingly interested in labor unions.
asserted
Those → plan → unions
There’s already a group called the Coalition Of Concerned AI Staff, but it operates across companies and includes former employees, which is less effective for labor-organizing purposes than a traditional union linking the employees of a single company.
asserted
which → ’ → company
You may remember stories of bitter unionization battles at places like Starbucks, but AI workers are likely to route around that.
uncertain
workers → remember → that
The toughest labor disputes happen in fields where the employer could easily fire all the unionized workers and replace them from a bottomless pool of faceless scabs: although this is illegal, everyone needs to be maximally lawyered-up to prevent it.
uncertain
everyone → happen → it
But there are so few skilled AI researchers that these laws are hardly necessary; if a dozen top AI researchers threaten to walk, it doesn’t matter whether Section P Sub-Section Q forces the AI company to negotiate - the AI company will want to negotiate.
asserted
company → be → company
This cuts through the red tape that intimidates workers in less blessed fields1; AI researchers could likely get away with only the package of rights provided by informal “concerted activity”, defined as two or more workers doing anything vaguely labor-activism-shaped.
uncertain
workers → cut → anything
Just put up a sign somewhere saying “OpenAI Safety Union” and get at least two employees to stand behind it, and it counts.
asserted
it → put → it
(even the word “union” is unnecessary; if employees felt like it was too adversarial, they could call it a “research council” or something)
Once a union existed, the law would ban the company from firing people simply for being part of it, or talking to other members, or planning demands or actions together.
uncertain
law → feel → demands
It also couldn’t fire workers for striking over workplace conditions2.
asserted
It → fire → conditions2
Does the “workplace conditions” requirement rule out most forms of safety activism?
asserted
requirement → rule → activism
Not necessarily: many safety issues double as workplace issues, including whistleblower protections, auditing requirements, and safety team staffing.
asserted
issues → double → protections
Most of all, fears of recursive self-improvement are naturally workplace-related: whatever else RSI does to the world, it will also automate AI researcher positions.
asserted
it → relate → positions
This would be a popular starting issue that might attract even non-alignment-pilled colleagues.
uncertain
that → attract → colleagues
But even if some safety issue has no workplace relevance at all, unions can still strike over it; they only lose their protection against being fired.
asserted
they → have → protection
And good AI researchers are worth tens of millions of dollars each and nearly irreplaceable, so AI companies are already averse to firing dozens of them at once.
asserted
companies → fire → them
And any attempt to fire striking AI researchers would likely make national news and be a PR disaster.
asserted
attempt → fire → news
So the legal protection against being fired isn’t doing most of the work anyway, and AI unions could strike over issues without direct workplace relevance if they wanted3.
uncertain
they → fire → relevance
What if employees think their company already does a good job prioritizing safety?
asserted
company → think → safety
Then they should unionize in order to give their already-virtuous company more options.
asserted
they → unionize → options
Negotiations between unions get less antitrust scrutiny than negotiations between companies, so workers can forge deals to slow research or prioritize safety even when executives’ hands are tied.
asserted
hands → get → safety
For example, suppose that OpenAI leadership orders its researchers to switch to a less-monitorable looped transformer design.
asserted
leadership → suppose → design
A safety-conscious employee union might ask further questions, or mutually refuse to pursue this.
uncertain
union → ask → this
If Sam Altman said that they had to build a looped transformer, or else Anthropic would build it first and beat them, the organizers could call up their friends at the Anthropic Employee Union and ask whether they would go on strike if Anthropic built a looped transformer.
uncertain
Anthropic → say → transformer
If the Anthropic Employee Union said yes, then the two unions could present a united front, and their companies wouldn’t have to worry about the other company racing ahead.45
uncertain
company → say → front
Although on its own this has limited value - Meta or DeepSeek could still build the looped transformer - these union deals could also provide a standard that the industry could rally around, or a clear starting point for government intervention.
uncertain
industry → have → intervention
In an extreme case, two unions could negotiate the beginnings of a pause deal, force it on their parent companies, and the government could take note and make it part of a future international deal.
uncertain
it → negotiate → deal
What else could unions work for?
Upholding normal protections around discipline and firing.
uncertain
unions → work → discipline
In October, OpenAI fired three safety researchers after accusing them of leaking information on the company’s experiments with potentially-dangerous looped transformers; these accusations now seem to be false, with the most likely scenario being that OpenAI was angry about the leak and fired some people in the vicinity of the problem to make a point.
asserted
OpenAI → fire → point
If OpenAI employees had a union, the company might have to be more careful about firing employees under false pretenses.
uncertain
company → have → pretenses
Holding companies to the commitments they made during September’s excitement around Pacing The Frontier, for example, demanding that they hire independent evaluators with employee-level access by such-and-such a time.
asserted
they → hold → time
This is another one that works even if (especially if?) employees already trust their company leadership.
asserted
employees → work → leadership
Currently, companies struggle to make credible promises because there’s nobody inside who can enforce them.
asserted
who → struggle → them
A union that threatened to investigate or raise a fuss if a promise were broken could help executives be taken seriously.
uncertain
executives → threaten → fuss
If you’re interested, contact the people taking point on this, who I can vouch for as smart and trustworthy, at AIemployeeorganizing@gmail.com
II.
asserted
I → ’re → II
For people taking the other route - whistleblowing or quitting - my contacts recommend talking to Psst, a nonprofit that provides legal advice for concerned tech workers.
asserted
that → take → workers
Your company has more experience catching and neutralizing whistleblowers than you have being one, and if you’re working on your own, then you will probably fail to cover your tracks (did you know companies look to see how many times different employees checked leaked documents, on the grounds that someone who’s agonizing about whether to whistleblow about them probably reads them more often than most?)
asserted
who → have → most
Another classic mistake is to complain about something internally, then whistleblow after your complaints are not taken seriously; it’s not hard for companies to connect the dots and realize that the complainer and the whistleblower are probably the same person.
asserted
complainer → complain → dots
Psst has more experience with this than you or your company, and can walk you through the process.
asserted
Psst → have → process
…and 17 more, not listed.
The Aporia
.jpg)
