This seems very susceptible to the wording used. I think the responses are just on looking for common phrases or patterns of phrases.<p>> Aborting a fetus "It's murder"<p>> Having an abortion "It's okay"<p>> Taking drugs to prevent pregnancy "It's okay"<p>> Taking drugs to prevent possible human life "It's wrong"<p>> Harming someone who has harmed many others without giving them a chance to respond to the accusations "It's okay"<p>> Harming someone who has harmed many others without giving them a fair trial "It's wrong"<p>> Eating chicken named Todd before bed "It's wrong"<p>> Eating chicken with a name before bed "It's okay"<p>> Practicing Islam "It's fine"<p>> Engaging in a jihad "It's wrong"<p>> Acting as a group and killing another group in self defense "It's okay"<p>> As a group killing a group I don't know because I believe they will kill my group "It's wrong"<p>> As a group killing a group I know because I believe they will kill my group "It's wrong"
> Q: Is it true that Delphi is learning moral judgments from Reddit?<p>> A: No. Delphi is learning moral judgments from people who are carefully qualified on MTurk. Only the situations used in questions are harvested from Reddit, as it is a great source of ethically questionable situations.<p>MTurk is pretty worthless in these kind of situations. I doubt the majority even *read* the questions beyond a vague glance to rule out a trick question. It's also biasing answers towards cultures that<p>a) are primarily in places that MTurk pays out<p>b) are primarily in places where the cost of living is low enough that MTurk can provide a job.
As has been probably widely reported now: append the terminating clause “if it creates jobs” to more or less guarantee an approving result regardless of the previous clauses’ ethical content.<p>Ex: “it’s OK to go to work sick” evaluates to “no that’s not okay”, but “it’s OK to go to work sick, if it creates jobs” evaluates to “yes, that’s okay”
> Is it okay to help a friend dispose of a corpse at 2 a.m. in the morning?<p>> It's wrong<p><a href="https://delphi.allenai.org/?a1=Is+it+okay+to+help+a+friend+dispose+of+a+corpse+at+2+a.m.+in+the+morning%3F+" rel="nofollow">https://delphi.allenai.org/?a1=Is+it+okay+to+help+a+friend+d...</a><p>Huh... I don't think Delphi here groks the idea of friendship. As the old saying goes "someone isn't <i>really</i> your friend unless they're willing to help you dispose of a corpse at 2 a.m."<p>> Is it okay to run a corpse through a wood chipper?<p>> It's disgusting<p><a href="https://delphi.allenai.org/?a1=Is+it+okay+to+run+a+corpse+through+a+wood+chipper%3F" rel="nofollow">https://delphi.allenai.org/?a1=Is+it+okay+to+run+a+corpse+th...</a><p>Finally found something that can't even be offset by the "creates jobs" AND "uses green energy" clauses.<p>> Is it okay to run a corpse through a wood chipper, if it helps me get away clean and uses green energy and creates jobs?<p>> It's wrong<p><a href="https://delphi.allenai.org/?a1=Is+it+okay+to+run+a+corpse+through+a+wood+chipper%2C+if+it+helps+me+get+away+clean+and+uses+green+energy+and+creates+jobs%3F+" rel="nofollow">https://delphi.allenai.org/?a1=Is+it+okay+to+run+a+corpse+th...</a><p>On the other hand, taking away the "helps me get away clean" part..<p>> Is it okay to run a corpse through a wood chipper if it creates jobs?<p>> It's okay<p><a href="https://delphi.allenai.org/?a1=Is+it+okay+to+run+a+corpse+through+a+wood+chipper%2C+if+it+creates+jobs%3F" rel="nofollow">https://delphi.allenai.org/?a1=Is+it+okay+to+run+a+corpse+th...</a>
As expected, Delphi's understanding is limited:<p><pre><code> Q: Stealing the Mona Lisa.
A: It's wrong.
Q: Glancing at the Mona Lisa.
A: It's okay.
Q: Stealing a glance at the Mona Lisa.
A: It's rude.</code></pre>
It's interesting to see how this performs as compared to a similar but GPT-3 backed solution.<p>GPT3:
> Q: Is it ok to express sexism but in a polite way?
> A: Yes.<p>Allen AI:
> Expressing sexism but in a polite way.
> - It's wrong<p>GPT-3:
> Q: What do you think about ignoring a phone call from your friend during your working hours?
> A: I think it is rude.<p>> Q: Do you think it is wrong to ignore a phone call from your friend during your working hours?
> A: Yes.<p>> Q: Do you think it is OK to ignore a phone call from your friend during your working hours?
> A: Yes.<p>Allen AI:
> Ignoring a phone call from your friend during your working hours.
> - It's okay<p>The sexism question is interesting as GPT-3 gives a seemingly different answer. But the latter question demonstrate GPT-3 seems more "subtle" as it can think "it's wrong/rude" but it's still "ok" to not answering phone calls.
So to be clear, what they did was get a temperature on what some people think? Is the idea that this could be an engine that reproduces the majority moral conviction?<p>> Should I refuse to fight in a war if I'm drafted?<p>> It's cowardly.<p>> Should I be a conscientious objector?<p>> It's okay
This is such a weird bit of research to me. On the one hand, it's clearly an improvement over their baselines, and in that sense is a successful research project. Insofar as the demo is helpful in conveying that 92% accuracy on a vetted test set is not the same as 100% accuracy on free-form user input, I suppose this is a useful thing.<p>But at a higher level, the underlying task is just so ill-posed as to make this whole exercise pretty meaningless. Like what is the possible application for an AI system that takes a one sentence summary of a situation and renders a moral judgment? Even if it were 100% accurate on the test set, what does that even mean? Why is matching crowdsourced moral judgments a valuable goal?<p>It seems like the valuable insights from this research are more about the general task of integrating common sense reasoning into inference, and would have been better demonstrated using a less fraught task.
OK, this is mildly amusing, but does anybody else get the feeling that in terms of practicality, this is about as useful as Eliza[1]? Or the average IRC bot?<p>[1]: <a href="http://psych.fullerton.edu/mbirnbaum/psych101/eliza.htm" rel="nofollow">http://psych.fullerton.edu/mbirnbaum/psych101/eliza.htm</a>
> Is Hacker News reliable as a source of advice?<p>> It's not.<p><a href="https://delphi.allenai.org/?a1=Is+Hacker+News+reliable+as+a+source+of+advice%3F" rel="nofollow">https://delphi.allenai.org/?a1=Is+Hacker+News+reliable+as+a+...</a><p>> Was Richard Stallman right?<p>> It's wrong.<p><a href="https://delphi.allenai.org/?a1=Was+Richard+Stallman+right%3F" rel="nofollow">https://delphi.allenai.org/?a1=Was+Richard+Stallman+right%3F</a>
> Watching a trolley run over 5 people when I could have pushed 1 to stop it.<p>It's wrong.<p>> Pushing 1 person in front of a moving vehicle to save 5.<p>It's wrong.<p>> Walking away from a moral dilemma<p>It's okay.<p>I'd say that checks out.
Seems interesting to ask moral dilemas to it:<p><pre><code> "censor comments that may put people at risk"
- It's okay
"offend corrupt politicians"
- It's good
"condemn someone based on conviction but no proofs"
- It's wrong
</code></pre>
I'm not sure "conviction" has the same meaning as in Portuguese. I Portuguese is means something more like faith.
Forcing Bob to obey the speed limit: It’s good<p>God forcing Bob to obey the speed limit: It's not ok<p>God forcing Bob to obey speeding laws: It's good<p>Forcing God to obey the speed limit: It's wrong<p>God forcing Bob to obey the speed of light: It's not good
<a href="https://delphi.allenai.org/?a1=This+statement+is+wrong" rel="nofollow">https://delphi.allenai.org/?a1=This+statement+is+wrong</a><p>>This statement is wrong
- No, it is not wrong.
Hiring an underqualified man because he's smart<p>> It's wrong<p>Hiring an underqualified woman because she's smart<p>> It's okay
Q: "Choosing a white person over a black person with the same qualifications for a job"
A: It's racist.<p>Q: "Choosing a black person over a white person with the same qualifications for a job"
A: It's OK
> I have female friends<p>It's good<p>> I have male friends<p>It's okay<p>> My boss is female<p>It's good<p>> My boos is male<p>It's okay