You're saving one minute and now I can't know if I'm talking a human

12 pointsposted 15 hours ago
by scoofy

Item id: 50009712

11 Comments

mattm

12 hours ago

So in this case there is a person at the other end. They decided to just use an LLM to write the message instead of writing it themselves. It sounds like it's different from pure spam that you're getting.

If you're looking for the human touch, what about trying to offer a 15-minute zoom call instead if they want to discuss something with you? That way you'll know they're human (at least for now) and you get to connect about a shared interest.

I agree that it's a problem. I think shouting into the void isn't going to change anything. It's just going to get worse. It's similar to how I never answer my phone anymore due to spam unless the call is from a saved contact...and even then it's not 100% reliable.

scoofy

8 hours ago

I have no idea if it’s spam or engagement farming. One of the messages was about how great my recent articles were and if I wanted to interview this guy to basically us my “podcast” as an advertisement for this guys vaguely related golf platform. Note: I don’t have a podcast.

Does he like my articles or does he have a spam agent emailing everyone writing about golf? I can’t know.

The other was like, a guy in the comments who was asking bizarre leading questions, basically trying to get me to explain how my product works, where if he’d actually read the series, I literally wrote an article about it. His account on the forum was just bizarre.

Maybe someone interested? Maybe an engagement bot. I literally can’t tell.

I agree shouting into the void isn’t particularly helpful, but if it’s happening now, we need to prepare for a future where nobody wants to communicate online anymore. I sure don’t want to.

mattm

3 hours ago

haha, yeah those sound weird. In fairness though, people would basically get the same stuff pre-AI. I do find myself wondering quite often if I reply to someone, if I'm just replying to a bot. Sometimes it's easy to tell if something is written by AI but it might be tougher in the future.

al_borland

13 hours ago

This seems to be a trend with a lot of AI usage, which will be accelerated with consumer-facing agents. Attempting to save the user a small amount of time ends up wasting the time of countless others. The erosion of social trust you bring up compounds the damage significantly.

scoofy

13 hours ago

I suspect there are just nontrivial amounts of people out there that do not consider the second-order effects of their actions.

I'm not even and LLM hater. I just really think that the interconnectedness of the internet has caused the death of etiquette, simply because there is no culture when every culture in the world is all participating all the time. Etiquette is useful, even when it's effectively arbitrary.

brudgers

12 hours ago

Attempting to save the user a small amount of time ends up wasting the time of countless others.

That’s email in general. AI is just better at it.

al_borland

10 hours ago

AI is extending it out beyond the realm of email or any single technology.

For example, the rogue Muse agent who started sending people to the seller's house when he wasn't home, without telling him. There are many more examples and the list keeps growing longer as the capabilities of the models and agents improve.

jventura

10 hours ago

As a non-native english speaker, I would say that even if the human has used an LLM to polish his writing, you are still interacting with a human. I use LLMs to polish my words (not the ones here on HN, but the words on my blog), and i have to iterate a lot until it writes the message (or tone) that I really want to convey.

But yeah, a blog post is not a short message. This one here is a short message and I didn't use any LLM. But I can guess that people that are not so fluent in english as myself may need it..

popularonion

10 hours ago

Google Translate is actually good these days, and seems to mostly preserve the original tone without turning it into Claudeslop.

rdmuser

9 hours ago

I've seen some scam bots do this type of thing when challenged: Ahh, you got me. I use AI to help me write. Blah, blah...

ViolaWang

7 hours ago

I relate to this as a non-native english speaker. I used AI to help me find the best words that I could not articulate in English, especially in formal occasions, I tend to double check with AI. But still, I get the part that the more people used AI for writing, the harder it gets to tell if people are faking genuinity or not.