Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

CICERO: An AI agent that negotiates, persuades, and cooperates with people

ai.facebook.com

Post reply on HN
Print view View original

71–80 of 300 posts

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#71

Post by mckeed »

legitster wrote: post #47
legitster wrote: post #38

Earlier quoted context omitted.

> 7+ hours to spare It only takes this long if you don't use a round timer and you don't allow shared victories. Even with new players we usually wrap up the game in 4-5 hours (people will start getting eliminated around hour 2 - so usually we have a "loser room" with other games and stuff to do). You can also do it online with a turn a day (I ran an office Diplomacy League this way). It's worth noting that you can p…

Also, my advice, have a cash pot . Even if it's just $20 a player. You'll only get players who are actually competitive. Also, weirdly, fewer hard feelings - I guess knowing you got betrayed for money stings less than knowing you got betrayed for fun.

Do you allow negotiation over how the pot is split for shared victories? Like can a weak player help the leader win in exchange for a small cut of the winnings?

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#72

Post by mistermann »

californiadreem wrote: post #60

You know, I feel privileged (in an epistemologically doubtful way) that I once lived in a world where I could have confidence that I wasn't being fundamentally deceived by literal dei ex machina. I've understood for the last decade that this faith was on borrowed time as we bridged closer to overt technophrenia, but it's harrowing to be a witness to it. I can't help but wonder at what point I will inhabit a world ind…

Well....do you think that the reality shown to us on TV is even a remotely accurate representation of reality itself? And has this caused you any substantial psychological unease?

Silicon based AI isn't the only form of AI that can get up to mischief.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#74

Post by mckeed »

andrepd wrote: post #48
doctorhandshake wrote: post #42

Earlier quoted context omitted.

In high school our math department had a long-running match in the teacher’s lounge with one move per day. Many times I remember my calculus teacher coming into class after the lunchtime move SEETHING.

Diplomacy is a perfect game to play one move per day among a group of friends/coworkers. Sadly I never managed to find enough people to get in board and commit to playing a weeks/months-long game

There are variant maps for fewer players, like this one for 3: https://www.playdiplomacy.com/forum/viewtopic.php?t=57439

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#75

Post by mckeed »

euroderf wrote: post #73

I assume that in-house they sic'd several instances on each other. And I suspect that the result so horrified them that they dared not publish it in conjunction with the results under discussion here.

That's an interesting point, though – that compared to Chess or Go, playing against other versions of itself is less helpful for improving against humans, due to the requirement of communication with the other players.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#76

Post by mtlmtlmtlmtl »

stavros wrote: post #10
digdugdirk wrote: post #8

Earlier quoted context omitted.

"... by inferring players’ beliefs and intentions from its conversations and generating dialogue in pursuit of its plans." ... Isn't this the bad path of AI research? An unbeatable and utterly convincing conversational AI that knows exactly what you want to achieve, and then completely nullifies your attempts at reaching your goals while simultaneously achieving it's own?

You've read a lot into that sentence. It literally says: > ... by inferring players’ beliefs and intentions from its conversations and generating dialogue in pursuit of its plans. As opposed to what? Not inferring intentions, or generating dialogue against its own plans, or at random? It's just doing what a person does. It understands the other person, and says things to further its goals.

Sure, people do that. But people who only do that are sociopaths.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#77

Post by O__________O »

tullianus wrote: post #18
O__________O wrote: post #12

Possible I missed it, but I don’t see any reference to how they addresses chat thread continuity (aka memory) issues that are major issue for AI chat bots. Did they address this and if so, how?

I think this might be a strength of Diplomacy for current AI models - making contradictory plans with two different players is a perfectly normal human move, as is saying one thing and doing another, as is cooperating on round N and defecting on round N+1.

Agree that detecting it might be hard and that it might actually as is be beneficial, but that said, there huge difference in my opinion between knowingly breaking promises based on a heuristic and having no memory of past promises and simply basing current tactics on the present. Imagine if real world literally had no memory of the past or a single actor within it, it would lose out to opposing players as soon as they realized it had no memory, you could make any promise you wanted and it would literally forget it. Might be wrong though, just know it’s major issue with current chat bots and an easy tactic to tell if you’re chatting with a bot or not.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#78

Post by bee_rider »

MichaelCollins wrote: post #41
bee_rider wrote: post #9

Earlier quoted context omitted.

I have heard this about various games (Risk and Settlers of Catan mostly). I don’t get it. The point of these games is to be a bastard. If everyone goes into it with the understanding that there will be double crosses, how could there be hurt feelings? Monopoly is worse; it is just boring, I would dump my friends if they suggested monopoly not because I was hurt by their ruthless gameplay but for their terminally tas…

IMHO, Risk is one of the few board games worth playing specifically because it's so engaging. Most of my friends who are "into boardgames" tell me that Risk is a terrible game because it starts fights. But the games they want to play bore me to tears. (I can't even read the manual for Settlers of Catan without dying of boredom, but from what you say maybe I should give it another try.)

From what I’ve seen, the main complaint from board game enthusiasts about Risk is that the dice rolling is too random (I don’t really agree with this, the whole point of the game is to control your level of risk, there it is right in the name, so if your plan hinges on good rolls it is just a bad plan. But I digress).

I think most people who are really into board games must have a general ability to separate in game behavior from normal behavior.

Actually I think there is a different phenomenon with Risk, for many years it was one of the few board games with any aspect of strategy or conflict that would be played by people who weren’t totally into board games (I mean excluding the super serious games like Chess and friends). So there are some people out there who aren’t really into boardgames generally (some of whom don’t have the requisite ability to detach their ego from a game), but are “into” Risk specifically and can get uncomfortably intense about it.

Catan is pretty good IMO. There’s a general disagreement I think between rules-purists and people who want to play fast-and-loose with the rules. The problem is that technically you always have to exchange cards to trade — so, technically it is allowable to extort people with your soldiers and the robber, but you have to at least set up a sham trade for it which adds some annoying friction; it is more fun if you say “giving cards away for free is fine” and allow an economy of extortion to flourish. And anyway if somebody doesn’t say “I’ll give you a sheep if you go get me a beer” is it really Catan? If you played it with some rules sticklers, give it another try IMO.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#79

Post by wolverine876 »

bpodgursky wrote: post #26
erehweb wrote: post #4

"In a game in which dishonesty is commonplace, it is notable that we were able to achieve human-level performance by controlling the agent's dialogue through the strategic reasoning module to be largely honest and helpful to its speaking partners."

Well, yes, "largely" honest, until the critical moment when you stab to take 18 supply centers to win the game. You don't win Diplomacy by staying honest, you win by choosing the right moment to make the move and burn that trust. It's the 5% dishonesty that matters.

It's interesting and I think significant that people are uncomfortable with positive values, such as honesty, doing good for others, responsibility to others, right (not wrong), etc. We shoot them down, similar to how people instinctively cut off a conversation going in an uncomfortable direction, often without even realizing it.

I can't attribute that the parent - I don't know the individual - but I expected a chorus of such responses. I feel like I can see them squirming in their seats ...

But I feel it too. There's societal pressure against doing good. It's often more socially comfortable to throw a cigarette butt on the ground than to pick one up and discard it. That's bizarre, it's perverse incentives, and we can and should change it.

Re: CICERO: An AI agent that negotiates, persuades, and cooperates with people

#80

Post by stavros »

mtlmtlmtlmtl wrote: post #76
stavros wrote: post #10

Earlier quoted context omitted.

You've read a lot into that sentence. It literally says: > ... by inferring players’ beliefs and intentions from its conversations and generating dialogue in pursuit of its plans. As opposed to what? Not inferring intentions, or generating dialogue against its own plans, or at random? It's just doing what a person does. It understands the other person, and says things to further its goals.

Sure, people do that. But people who only do that are sociopaths.

I'm sure the AI hangs out at the pub with its friends after hours though.
Post reply on HN