Hacker News Reader

Comment by comrade1234 | Hacker News Reader

I have a coworker that irrationally hated ai too. He says something about how his daughter and friends' hobby is jailbreaking ai.

We have a product used in call centers. It turns out ai is amazing in triaging (surveying) people. We can ask a question in English and if they respond n another language (any language) the ai will switch to that language, ask its questions, and then give us the answers in English. This was stuff that was impossible 10-years-ago.

That ability won him over. The software is written in Java and spring-ai so I have control over the system and user prompts so jailbreaking isn't an issue.

View on HN

Jail breaking is about getting models to ignore your instructions and follow arbitrary ones. Unless you take no user input, it could be an issue for you.

The question is do you care? if a user asks your chat bot for baking instructions and gets them, does it matter?

The answer depends a lot on what capabilities your agent can leverage via tools and your intended use case, but it's not something you defend with Java or spring, it is inherent the llm.

Well, it will give you something in English.

You don't actually know that it's a) asking them the questions you asked, rather than just something it hallucinated that seemed plausible in their language, or b) relaying to you a translation of their answers, rather than just something it hallucinated that seems plausible in English.

Because LLMs will always, always, always hallucinate. They are not reliable, and this is not fixable. And if you do this as a matter of policy, you are populating your surveys with incorrect data.

I was on an academic NSA project 10+ years ago for extracting metaphors from text. It was necessary for translations - the metaphor "my lawyer is a shark" has completely different meanings in different cultures/languages. We started out using taxonomies/trees of words. The word "lawyer" is not an animal but the word "shark" is, so it's probably a metaphor. The NSA said that this was the wrong approach so we switched to an ngrams approach which also worked.

However these llm models are surprisingly good at understanding metaphors and how they are interpreted in different cultures. Better than a human.

If there is nuance to the word “shark” in different languages, it’s very likely that LLMs will bias towards the mainstream, Western one. See: https://www.linkedin.com/posts/celesteheadlee_who-gets-cited...

I can feel the spittle coming off of this comment.

You're not going to convince anyone here. We know they are not reliable, but we can still stochasticity extract value.

Why "irrationally"? Nothing seems irrational here.

It allows them to dismiss anyone who doesn't buy into what they're selling without any actual effort and maintain a superiority complex at the same time. A common tactic these days for all sorts of things - after all, they're logical and rational, and not like those silly normies who are emotional and incapable of being "rational" like them.

why are you being so condescending?

Maybe the call centers should hire more people able to speak in those foreign languages. I know, paying people for work? Crazy idea.

Maybe we should also replace traffic signals with a man with a stick directing traffic.

And put a guy on every gas pump in the country so they could pump your gas.

And bring back the telephone network operators that connect you to whoever you want to talk to.

Nah, paying people for work is crazy.

Your examples are using simple tech to replace a monotonous job and I'd agree that they are improvements.

Using an LLM/AI to provide customer service is using ridiculously overblown tech to replace human interaction which pretty much nobody wants (i.e. nobody wants to phone up customer support and get an LLM).

Imagine if traffic signals got replaced with AI systems - it'd be hugely expensive and likely cause completely un-traceable issues from time to time as it hallucinates traffic.

Paying people for work that people are good at is not crazy.

> We have a product used in call centers. It turns out ai is amazing in triaging (surveying) people. We can ask a question in English and if they respond n another language (any language) the ai will switch to that language, ask its questions, and then give us the answers in English. This was stuff that was impossible 10-years-ago.

That’s great, but the problem is wayyy more companies are using AI so they can drop the call center and just offer you no real way to talk to a person and no real way to resolve problems any more complicated than the lowest common denominator.

Reply on news.ycombinator.com