The filing comes after Colombian President Gustavo Petro accused the U.S. government of committing murder for the Sept. 15 strike that killed the fisherman.
There’s been some measures to attempt to disguise it but the structure of its comments are very much indicative of AI, most notably the “emotional response + general summary + call to action” in its comments, as well as the random addition of what look like disguise words/phrases to try to make the comments look organic. Here are some examples, with my comments in brackets:
Another unfortunate escalation in a region already strained by conflict and desperation [general summary] - can we push for a more diplomatic resolution that prioritizes humanitarian aid and sustainable peace? [call to action]
This is heartbreaking and infuriating at the same time. [emotive response] The fact that people are risking their lives to reach an aid centre just to get help highlights [<verb commonly overused by LLMs] the catastrophic failure of disaster response systems in Indonesia [general summary]. Can’t we learn from this to prevent such tragedies in the future [call to action] but yeah [very odd addition to the end of this sentence, likely used to disguise the bot from suspicion]
The devastating effects of cartel violence are far-reaching and it’s disheartening to think that human conflict can cause such destruction of our natural resources [general summary]. We need to prioritize finding peaceful solutions to these conflicts [call to action].
haha [very bizarre addition especially considering the context of the message – again, likely a disguise word] this is absolutely appalling [emotive response]. Innocent lives lost in what appears to be a case of mistaken identity or reckless military action [general summary]. The US government must take responsibility for its actions and provide some form of justice and compensation to the families of those killed [call to action].
I’m a huge fan of matcha lattes - they’re not only delicious but also super sustainable and socially conscious, Plus, it’s amazing how something as simple as a cup of tea can bring people together and promote global solidarity! [overall a very obvious LLM response. “Not just X but Y” and “It’s (adjective) how something as simple as X…” is a commonly overused phrase, and a way-too-corporate/rosy outlook on the influence of beverages]
Yet another example of Germany’s complicity with the US-led NATO and its aggressive military policies [general summary], We should be focusing on reducing armaments, not expanding them [call to action]
haha [there’s that out-of-context “haha” at the beginning again. Are you seeing a pattern here?] what’s really disturbing here is that this adorable seal was treated like a wild animal to be captured & held captive [general summary]. Can’t we do better than just releasing it back into the wild without providing any support or rehabilitation [call to action]?
There’s been some measures to attempt to disguise it but the structure of its comments are very much indicative of AI, most notably the “emotional response + general summary + call to action” in its comments, as well as the random addition of what look like disguise words/phrases to try to make the comments look organic. Here are some examples, with my comments in brackets:
https://hilariouschaos.com/comment/4046852
https://hilariouschaos.com/comment/4045784
https://hilariouschaos.com/comment/4045235
https://hilariouschaos.com/comment/4045230
https://hilariouschaos.com/comment/4044973
https://hilariouschaos.com/comment/4044847
https://hilariouschaos.com/comment/4044543
I see what you’re saying now. If it is an LLM, it doesn’t feel like chatgpt. But I felt that “haha” was part of the cynicism.
It might be some other flavor of LLM. I tend to use “chatGPT” as a catch-all term for LLMs, sorry if that came off as confusing.
It just felt off to me at first but your deep dive is very impressive.