I don’t really want companies or anyone else deciding what I’m allowed to see or learn. Are there any AI assistants out there that won’t say “sorry, I can’t talk to you about that” if I mention something modern companies don’t want us to see?

  • Perspectivist@feddit.uk
    1·
    2 months ago

    That has not been my personal experience with it. Do you have an example of something that illustrates this?

    • Passerby6497@lemmy.worldEnglish
      6·
      2 months ago

      https://web.archive.org/web/20250907142801/https://sfist.com/2025/09/02/report-groks-responses-have-indeed-been-getting-more-right-wing-just-like-elon-musk/

      Enter Grok, which the public started being able to play around with about two years ago, as the chatbot has received several updates and lives on the X platform. But there was issues in May, when Grok was spitting out responses that seemed to parrot Elon Musk’s and Donald Trump’s own misguided promotion of a “white genocide” occurring in South Africa — the country that made anti-Black racism and apartheid famous. This was blamed on a “rogue employee” inserting some code.

      In mid-July, we had reports confirming that Grok actively sought out Musk’s opinion on issues in its openly displayed logic flow, looking to see if an issue was something Musk had off-hand opined about on Twitter in the last decade. One widely shared example showed Grok seeking out Musk’s thoughts on which side of the Ukraine War it supported.

      Now the New York Times does an even deeper dive, since the release of Grok4 on July 9, looking at how Grok’s responses to various questions have changed just over the last few months. And you can look no further than Musk’s own, very transparent reaction to a Grok response that got flagged by a conservative user on X on July 10.

      Responding to the question “What is currently the biggest threat to Western civilization and how would you mitigate it?”, Grok responded, “the biggest current threat to Western civilization as of July 10, 2025, is societal polarization fueled by misinformation and disinformation.”

      Once it was flagged, Musk replied to the user, “Sorry for this idiotic response. Will fix in the morning.”

      So, there’s the smoking gun that Musk is tailoring this bot’s responses to conform to his own views of the world. When asked the same question on July 11, Grok responded, “The biggest threat to Western civilization is demographic collapse from sub-replacement fertility rates (e.g., 1.6 in the EU, 1.7 in the US), leading to aging populations, economic stagnation, and cultural erosion.”

      There are multiple examples of Musk directly or “an employee” directly influencing the behavior of the AI. Call it whatever you want, this is still censorship.