Maybe we need a better-scaling legal system that does not take years to resolve simple disputes. I am not convinced AI can replace lawyers, but more people having access to the law without extremely costly fees does sound good to me.
Usually the concerns I've seen are around the lines of "the experts are gonna favor the powerful incumbents who they see a lot" buttttt if our system is already doing that AND costing way more anyway, barring entry entirely to many, is that so much worse? Even assuming we can't try to regulate that?
Typically, the arbitrator is very well versed in the rules/area they are enforcing, and can see right through a snow job. In my province there is a tenancy tribunal that has to make a ruling before a court gets involved, and the courts rarely disagree with a decision made by the arbitrator in the 1% of cases that do get appealed. Because the rules are so clear, you can shovel piles of paperwork at them, and they will ignore it because they know what pieces of paper they actually need to make their decision.
It isn't for complex big money cases, but it means that courts have more time for those cases because they aren't dealing with ticky tacky minor disputes over damage deposits and fender benders.
Tenancy arbitration is just one of them, BTW. Just about every industry or agency has an outside ombudsperson that is the first step of dispute resolution before a court will be interested.
If anything, some of the largest civil law countries in Europe, like Germany and Italy, can arguably be considered worse than common law countries in all of these respects. Whether that's incidental or due to civil law, I cannot say.
https://worldpopulationreview.com/country-rankings/lawyers-p...
Apparently a highly biased measurement because the German legal system has more public funding compared to the US. I asked «For an argument against litigious societies, where someone points to "lawyers per capita" to compare the US with Germany, what might be a better yardstick?»
The response was interesting to me: the key bit was "Germany has a massive civil service and specialized court system where many law degree holders work as judges or state administrators". I'm a geek from New Zealand so unfortunately that isn't something I can validate from knowledge.
The English legal system/philosophy (including jury trial and the adversarial prosecution/defense dichotomy) derives historically from English/Anglo-Saxon traditions (the 'common' law of England) and differs substantially in some respects from the legal systems in other countries which derive from Roman and later French/Napoleonic conventions (known confusingly as 'civil' law).
EDIT: although you mention civil law in your other comment - what I am getting at is the difference in role of judge/prosecutor in criminal cases between the two systems being comparable to GGP's suggestion
I'll probably yield to my temptations eventually and proceed anyways. Lord help me from all the creative but completely detached interpretations I'll land on.
https://ice.dhs.dev/program/13732-human-trafficking-investig...
https://atf.doj.dev/program/44825-open-gun-store-need-ffl
LMK if you want to know more.
I saw a sequence diagram browsing around, seemed to be specific to a sample scenario?
Fair question, and the "training course" read is not an accident; it's the same shape underneath. A program is an ordered chain of modules, each with a declared actor and typed inputs/outputs. Courses are also that. So it renders with the same components. The sequence diagram you found isn't a sample scenario, it's the deal template's actual step graph; the thing an instance runs on.
Three authored files per domain:
- an ontology: the domain's vocabulary, its regulatory frameworks with real citations, the O*NET occupations that staff it, the systems of record it touches
- intents: what a person actually shows up wanting ("open a gun store, need an FFL"), with typed parameters
- deal templates, one per intent: ordered pipeline_steps, each with an actor, inputs, outputs, and a policy_check
The page you clicked is generated from the last two deterministically. No model in that path.
The part that speaks to your TLA+ instinct: I deliberately don't formalize what the law means. I formalize the procedure, and bind each step to the provision that authorizes it. Formalizing semantics is exactly where you get the creative, detached interpretations you're worried about, because every gap gets filled by the model's guess. Formalizing procedure asks the model to transcribe and cite, which is checkable:
- every step input is a ref; param:x, step:3.some_output, system:NICS.event; and it has to resolve. A step: ref must name an earlier step's declared output, so the dataflow is a DAG with referential integrity.
- every step's policy_check must name a framework declared in the ontology. A step that no provision authorizes fails validation.
So most hallucination becomes a build error instead of a plausible sentence. That's the whole trick. Not a smarter model; a narrower artifact.
Concretely, since you're right to expect slop: my first pass at four new agencies came back with 100% of step inputs referencing parameters that didn't exist, and prompts that literally said "Subject?". The validator refused all forty programs. That's the mechanism working; I'd have merged them on a read-through.
Intents and flows for ATF, if you want to see the layer under the program page: https://wiki.doj.dev/agent/atf
Limits, since you'll ask. It decides nothing; no adjudication, and consequential steps are human-gated. It's also not a formal method: the invariants are referential integrity and citation binding, not model checking. The genuinely temporal parts are the deadlines, and those do bite; the NLRB's six-month charge window runs from filing and service, with service being the filer's own duty, so a filing-date-only clock computes the wrong date on a deadline that destroys the claim if you miss it.
Re: the sibling comment about discretion; that's the actual pitch. Discretion hides in the gap between the written rule and the practiced procedure. Writing the practiced procedure down, with a citation per step, is what makes the gap visible.
Each "program" here is a government program, agents orchestrate everything including the collaboration between all parties required.
High points: I have been able to help over 100 people get housing with no HITL on my side.
Note: Each host/subdomain is a project, they all inherit policy from each other and that drives the program generation and orchestration layer. Policies can be managed for the diff agencies at rnc/dnc.dev
This fee should help pay for the courts and reimburse and employer for time spent defending itself.
(With the judge / magistrate able to wave the fee on compassionate grounds)
s/nuance/nuisance/It’s also on the court system to leverage procedure correctly to add safeguards against the kind of behaviour you describe.
I understand anecdotally the US system is more open to abuse of this nature, but I don’t know the details. This article though was specific to UK.
Of course it does, often. But that's a pathology, not a feature to be proud of.
It's true that some complaints are irrational, vindictive, or vexatious, but the system seems quite good at identifying those already.
If the complaint has a high probability of success then it has a low probability of making the plaintiff pay costs. If it has a low probability of success then isn't that what we're trying to deter?
When the odds of success are 50/50, in this context that's a nuisance complaint because making it "free" to turn all of those into real adjudications would flood the system with exactly the cases where the outcome is unclear which are correspondingly the hardest to efficiently resolve. The only practical way to address that is to clarify and simplify the rules to begin with so that the outcomes are predictable rather than ambiguous, and then we're back to not really deterring anyone because it's obvious who is going to win.
And in the meantime plugging up the system with unlimited questionable complaints isn't very prudent.
1% might not deter, but I bet it still deters more than you want. 10% will deter a lot despite those being cases that should happen. Especially because these non-experts don't know the actual odds. Their fuzzy guessing range is going to include worrying amounts of failure chance even for very strong cases.
But then won't experts be offering that insurance on easy to win cases left and right?
> 1% might not deter, but I bet it still deters more than you want.
You don't actually want to deter meritorious cases at all, but we don't have that as an option. You can either have more false negatives or more false positives. Requiring either one to be zero is going to make the other one unsustainably large.
And that's the case even if there isn't any formal penalty for filing a losing case, because there is still a time cost. This is actually the "problem" AI is creating -- the time cost still isn't zero but now it's lower and therefore less able to serve its previous role of deterring frivolous cases.
> Especially because these non-experts don't know the actual odds.
Having something that can tell them that is essentially required regardless or they'll be continuously initiating cases with a <1% chance of success.
No, I don't think so. Too much research for not enough profit.
> And that's the case even if there isn't any formal penalty for filing a losing case, because there is still a time cost.
Time is cheap when your wages are low, while penalties are not cheap.
> Having something that can tell them that is essentially required regardless
It's not a binary between knowing the odds accurately or not at all. Even with good advice there will still be uncertainty.
Isn't this basically how things work now when lawyers take cases on contingency?
> Time is cheap when your wages are low
Disposable time is very expensive when your wages are low. You have to work more hours just to make rent so you have less of it to begin with and then you have to spend what little you have left doing all the things you can't afford to pay someone else to do.
> It's not a binary between knowing the odds accurately or not at all. Even with good advice there will still be uncertainty.
If you know that the odds of winning are between 5% and 100% then you barely know anything and the result of proceeding with only that information would be a huge number of losing cases. If you know that they're between 90% and 100% then that's the situation where an insurer should see that and want to make money.
If there's a huge payout, sure. They're going to ignore you for a $10,000 case.
> Disposable time is very expensive when your wages are low. You have to work more hours [...]
That's only if you can get a ton of hours. A lot of people can't.
> If you know that they're between 90% and 100% then that's the situation where an insurer should see that and want to make money.
I'll just say it again: Too much research.
Insurers hate custom stuff and charge tons of money for it. If you have little money to spare, the fee just to get an insurance company to analyze your case is already too much for you to go through with it.
The entire system basically doesn't work for a $10,000 case. That's probably about the worst case scenario. For small claims you can have an abbreviated process because nobody is that worried about always getting it right when the amount in question is $400. When it's a million dollars then it's worth the time to go through the whole process. But $10,000 is enough that you can't just dismiss accuracy concerns yet not enough to do the things needed to actually address them.
For example, you can get into federal court because you have a dispute between residents of different states, but only if the amount in controversy is also at least $75,000. Less than that and the federal courts won't even look at it.
> That's only if you can get a ton of hours. A lot of people can't.
That's when they have to take a second job or moonlight driving for Uber etc.
> Insurers hate custom stuff and charge tons of money for it.
Insurers hate bespoke stuff because their business is about pooling common risks. If you want to insure your Bengal tiger against the risk of contracting a tropical disease while being transported through Brazil then they're going to quote you the "get lost" rate.
Common legal risks are often pretty predictable and they wouldn't necessarily even have to do much analysis. List the elements the claimant would have to prove in order to win and have the insured assert that they're true and specify how they intend to prove it. If they lie to the insurance company and that comes out in the proceeding then the policy is void and liars get what they paid for. If they told the truth then they have a predictably high probability of winning and that's the sort of risk insurance companies like.
For instance if you know for a fact that you have been fired due to your race (perhaps you were outright told this is the case), but have no evidence (you weren't recording and there were no witnesses), your case has merit (it really did happen), but unless you get lucky and some evidence appears in discovery your odds of success are miniscule.
I'd argue that in such a case preventing such a person from seeking justice would be wrong, since they were really victims and diserve redress.
If the system isn't actually going to give them redress then what benefit is there in consuming the resources of the complainant and the court only to end up with the same result? The problem you have to solve in those cases is how to make it so someone with a meritorious claim would have a high probability of winning.
that would encourage lawsuits against the state and big corporations, and make it less worth it to sue small businesses or your own employees. thats common sense because bigger more powerful organizations tend to be more evil (as in breaking the law because they can afford to pay fines) and money is a good measure of power in capitalism.
A lawyer isn’t going to sign on to take liability risk unless they spend quite a lot of time analyzing the AI’s outputs for possible blunders?
So it structurally can’t cost significantly less.