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

101–110 of 300 posts

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

#101

Post by irq-1 »

elijaht wrote: post #19
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…

I would say Diplomacy specifically has features that make it cause hurt feelings almost intrinsically 1. It's a long game- after hours of play, getting doubled crossed unexpectedly sucks because of how much you have invested 2. The game almost purely relies on cooperation/other players. So double crossing someone really screws them- it almost certainly means I have no recource (luck, dice rolls, individual tactics).…

Uniquely there isn't anything in the game, or on the board, to negotiate other than cooperation. You're not trading corn, or loaning money, or anything. Even "positioning" of pieces doesn't matter because you need allies to do anything, and you'll lose any position without allies; you can't build a fort or get a bonus by holding Indonesia. It's just 'please trust me' and 'please be trustworthy'.

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

#102

Post by bo1024 »

pesenti wrote: post #20

Paper: https://www.science.org/doi/10.1126/science.ade9097 Code: https://github.com/facebookresearch/diplomacy_cicero Site: https://ai.facebook.com/research/cicero/ Expert player vs. Cicero AI: https://www.youtube.com/watch?v=u5192bvUS7k RFP: https://ai.facebook.com/research/request-for-proposal/toward... The most interesting anecdote I heard from the team: "during the tournament dozens of human players never even su…

Hmm, I guess facebook doesn't have to go through IRB for human subject experiments, nor does Science require it, apparently.

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

#103

Post by irq-1 »

mikelevins wrote: post #66
tullianus wrote: post #15

Earlier quoted context omitted.

Two features that make Diplomacy particularly bad: 1. Length. An in-person game can last easily 12 hours, and is mentally exhausting. Getting 10 hours in and THEN getting screwed by your friend feels worse than a game with less time investment. 2. Design. It's difficult to survive the opening few hours without alliances, but only one player can win, so everyone is incentivized to defect at exactly the moment they thi…

I was a member of an enthusiastic friends-and-family gaming group through the early and middle 1980s. We spent a good chunk of that time on Diplomacy, then moved to modified Diplomacy on custom maps (my first wife worked for USGS Map Sales, and so we had good maps to work with in creating our own Diplomacy maps), and from there we graduated to Nomic. If you think Diplomacy is long and exhausting, wait till you get a…

This? https://boardgamegeek.com/boardgame/14451/nomic

> In the words of Nomic's author, Peter Suber: "Nomic is a game in which changing the rules is a move. In that respect it differs from almost every other game. The primary activity of Nomic is proposing changes in the rules, debating the wisdom of changing them in that way, voting on the changes, deciding what can and cannot be done afterwards, and doing it. Even this core of the game, of course, can be changed."

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

#104

Post by californiadreem »

mistermann wrote: post #95
californiadreem wrote: post #91

Earlier quoted context omitted.

Humans deceiving humans via technology is a tale as old as time. Plato imagined human torch-bearers manipulating those chained in the cave. He didn't imagine machines doing it autonomously. Incarnate technology deceiving humans is a different domain entirely. Human motivations are ultimately comprehensible by other humans. But what do I know of the motivations of (for example) the Bible? What happens when the living…

> Humans deceiving humans via technology is a tale as old as time. Plato imagined human torch-bearers manipulating those chained in the cave. He didn't imagine machines doing it autonomously. And despite a substantial subset of the population knowing this, we continue to do nothing to address it - if anything, more people are devoted to giving people even more powers to deceive at massive scale. > Incarnate technolog…

I don't think we disagree at all. The empire never ended, I guess.

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

#105

Post by worldsayshi »

EGreg wrote: post #49
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?

Just wait until we have AI swarms. So far all this has been with one player, amid others, no collusion. You’re going to be surrounded very shortly by sleeper bots, including on HN. Relying on dang and others to root out bots will be futile. A swarm could easily collude to downvote people or get them ostrasized by their own friend group, as we have already seen when it came to crypto, metoo, BLM, lockdowns, vaccines a…

There are ways to ensure that people are people on the internet though. Not good ways but there are ways.

You could for example limit users to those that log in with electronic id's issued by a government or other organisation that you trust to assert that the user is human and then force real names or a single user name for that e-id.

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

#106

Post by SamBam »

ghaff wrote: post #100
munk-a wrote: post #98

Earlier quoted context omitted.

Binding or non-binding treaties? Because in diplomacy everyone is always making non-binding treaties, promises and threats - and that's where the betrayal comes in. I have heard of some people using binding treaties in diplomacy (i.e. three turn ceasefire between France and Prussia) but that has always struck me as a terrible idea because you're shifting moves into the open and things get boring and cliquish.

I think I've played both ways (or at least played in a group where people didn't break treaties even if they could) but it's been years.

A "binding" treaty where the terms of the treaty are supposedly binding, but, like on the real world stage, the only penalties for breaking them is a lowered trust from all the other countries/players, sounds fine.

A binding treaty where if you break them someone will say "that's against the rules of the game, and the rest of us will stop playing if you don't comply" sounds terrible.

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

#107

Post by SamBam »

legitster wrote: post #27
tullianus wrote: post #15

Earlier quoted context omitted.

Two features that make Diplomacy particularly bad: 1. Length. An in-person game can last easily 12 hours, and is mentally exhausting. Getting 10 hours in and THEN getting screwed by your friend feels worse than a game with less time investment. 2. Design. It's difficult to survive the opening few hours without alliances, but only one player can win, so everyone is incentivized to defect at exactly the moment they thi…

> An in-person game can last easily 12 hours How are you playing it? We've never had a game run over 4 hours.

I've generally heard of people playing over a weekend. So things like meals and sleeping sometimes happen during the negotiation turns.

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

#108

Post by machiaweliczny »

gcanyon wrote: post #87

It seems like with this, Meta is doubling down on the Facebook platform's use as a dark influencer in politics. It's bad for humanity to enable advanced profiling and general targeting of demographics, but it's a whole other level if the communication itself can be targeted at the individual level. We're headed for a reality where I don't know if my friends' friends are real or not, and they each entice me with argum…

This is already done but more subtle. Probably the reason why Elon bought twitter.

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

#109

Post by mattnewton »

bee_rider wrote: post #9
kneebonian wrote: post #5

For those hearing of this game for the first time and think it sounds fun. Be warned, it is considered to be a game that will end friendships that have lasted for years, and monopoly is considered to be a game that causes less bad feelings. An AI may be able to speak like a person but will never be able to hang onto that long burning simmering hatred from when Brad didn't support my army and instead flipped on me by…

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…

Diplomacy is quite a bit different because you 100% win by being the nicest asshole. Past a small mechanical threshold, there really is no way to make it up in strategy or luck, it is only the interpersonal relationships and convincing that wins games. That, coupled with the fact that the game takes a long time and encourages a lot of communication, means the recipe to get invested and blur the boundary between game actions and lies people told you to your face and got you to trust them, can be very unlike risk. Finding out your friends can lie and manipulate you can be like finding out Santa clause isn't real.

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

#110

Post by ilaksh »

gcanyon wrote: post #87

It seems like with this, Meta is doubling down on the Facebook platform's use as a dark influencer in politics. It's bad for humanity to enable advanced profiling and general targeting of demographics, but it's a whole other level if the communication itself can be targeted at the individual level. We're headed for a reality where I don't know if my friends' friends are real or not, and they each entice me with argum…

Like I mentioned on Twitter the other day, this is the age of mass information manipulation, biowarfare, and 5,000 ton orbital rockets.

Maybe the AIs will eventually get smart enough to save us from ourselves.

Post reply on HN