[kgdiem]

Genuine question, how could Claude have been used for the military action in Venezuela and how could ChatGPT be used for autonomous weapons? Are they arguing about staffers being able to use an LLM to write an email or translate from Arabic to English?

There are far more boring, faster, commodified “AI” systems that I can see as being helpful in autonomous weapons or military operations like image recognition and transcription. Is OpenAI going to resell whisper for a billion dollars?

[janalsncm]

You can’t embed Claude in a drone. You could tell Claude code to write a training harness to build an autonomous targeting model which you could embed in a drone.

[kgdiem]

Fair. Didn’t think the DoW did much R&D or manufacturing. Would think the standoff would be with Anduril, Northrop, Boeing, Booze, etc.

[lyu07282]

Do you not have any imagination?

Who is going to read the whisper transcripts of mass surveillance to make decisions on who to target for repression? That's what LLMs are good for, it allows mass surveillance to scale. You can feed it the transcript from millions of flock cameras (yes they have highly sensitive microphones) for example. Or you hack or supply chain compromise smartphones at scale and then covertly record millions of people. The LLM can then sift through the transcripts and flag regime critical language, your ideological enemies or just to collect compromat at scale. The possibilities are endless!

For targeting it's also useful, because you want to indiscriminately destroy a group of people you still need to decide why a hospital or school full of children should be targeted by a drone, if a human has to make that decision it gets a bit dicy, people have morals and are accountable legally (in theory), if you leave the decision up to an AI nobody is at fault, it serves as a further separation from the violence you commit, just like how drone warfare has made mass murder less personal.

The other factor is the amount of targets you select, for each target you might be required to write lengthy justifications, analysis on collateral damage and why that's acceptable etc. You don't want to scrap those rules because that's bad optics. But that still leaves you with the problem of scalability, how do you scale your mass murder when you have to go through this lengthy process for each target? So again AI can help there, you just feed it POIs from a map with some GPS metadata surveillance and tell it to give you 1500 targets for today with all the paperwork generated for you.

It's not theoretical, that's what Israel did in their genocide of the Palestinians, "the most moral army“ "the only democracy in the middle east":

https://en.wikipedia.org/wiki/AI-assisted_targeting_in_the_G...

And here is the best part: none of this has to actually work 100%, because who cares of you accidentally harm the wrong person, at scale, the 20% errors are just acceptable collateral damage.

[kgdiem]

If I was building this in a system design interview I would use whisper, NLP, and “classic ML” classifiers with deterministic results. I would not want an LLM in the loop at all. Facebook and Google have been able to target you better than you could even perceive for years.

LLMs are slow, expensive and inconsistent. More importantly It’s not the right tool for the job.

Really feels like more “oohhh look at how important and scary LLMs are”.

*edit* PS, my company does marketing, communication and trade surveillance for FINRA registered broker dealer firms. If the CCO or anyone else with admin access wanted to monitor for someone talking badly about them they absolutely could update their list. No LLMs in the loop, very scalable, affordable, auditable and reliable. LLMs are just an interface not a solution for analysis.

[Otterly99]

I was asking myself the same question this morning and kind of came to the same conclusion as you. It makes much more sense to design a automated drone piloting system built with a decision-making algorithm than use a LLM. For Mass surveillance, I can see a bit more the use case, where you can use classical methods to process the information about a specific person but use the LLM to generate summaries or synthesize it when you have information that is not too organized or comes from different sources. I definitely think there is some overconfidence on the side of the decision makers in the government on what these tools can be used for. Also maybe some wishful thinking on what LLMs will be able to do in a few years?