3 ms·

and? All of the AI providers intentionally introduce biases: https://openai.com/global-affairs/introducing-openai-for-government/ https://openai.com/global-aff

by itake 1y ago

and? All of the AI providers intentionally introduce biases: https://openai.com/global-affairs/introducing-openai-for-government/ https://openai.com/global-affairs/introducing-openai-for-gov... https://www.anthropic.com/research/evaluating-feature-steering https://www.anthropic.com/research/evaluating-feature-steeri...

unshavedyak 1y ago
It is pretty interesting that this model will have two forms of bias though. One model derived from the company perspective and its training data, and two from Elon himself. Months ago this model would have promoted Trump, but now it'll call Trump disastrous for the economy. I don't know what to think of general company biases, and we've all been expecting biases to start favoring share holders eventually.. but biases based on twitter rants potentially changing day to day certainly is a new unique feature of Grok i guess.
- czl 1y ago
  > Months ago this model would have promoted Trump, but now it'll call Trump disastrous for the economy. There’s a well-known quote often attributed to economist John Maynard Keynes: “When the facts change, I change my mind. What do you do?”
  - unshavedyak 1y ago
    Yea but this example wasn't of the facts changing. It was of opinions changing. Musk had one opinion months ago, and a different one now. The bot explicitly searched twitter for the current opinions of someone with a history of less than stable actions (i'm trying to be generous for sake of neutrality). This bot isn't at the whims of the corporate oversight or advertisements, or even direction of a CEO - the bot is at the whims of every post from a chronically online and addicted Twitter personality. Even if you ignore Elons other flaws, making a bot the sum total of everything he's said on Twitter is pretty impressive. .. and not in a good way.
    - czl 1y ago
      This isn’t just about opinions randomly changing. Musk’s views seem to have shifted in response to changing facts -- like the passage of a major spending bill, whichtie directly into his concerns about government debt. So it's not surprising that his stance evolved over time. The way the bot reflects his views is a bit awkward, I agree. I assume it’s something the Grok team will want to improve. One possible reason for the behavior is that if Twitter data was used during training, and Musk is a dominant voice there, the model may have learnedto rely heavily on his posts -- especially if that helped it predict or score better during training. To Musk’s credit,he’s pushed for more transparency. That at least lets people see these odd behaviors and raise questions. With other models, like OpenAI’s or Anthropic’s, it’s harder to tell when something similar might be happening. And while Musk definitely says someoff-the-wall things, his track record overall is hard to ignore-- whether one agrees with him or not.
spacechild1 1y ago
There is a slight difference between feature steering and intentionally installing the (de-facto) CEO as the principal source of truth.
- itake 1y ago
  Keep going. I thought Anthropic’s CEO is the source of truth that AI based on his belief that it should avoid these topics. Musk has different opinions than Dario, but they are both introducing biases into their respective companies
  - nerevarthelame 1y ago
    Choosing not to answer - regardless of whether or not that was a rule mandated by the CEO (an unsourced and unlikely claim given the corporate structure of most large organizations) - is far different than insisting on an answer from whatever the CEO last decided to tweet. One is returning "null." The other is not. One says, "Figure that one out yourself." The other says, "Here is the truth."
    - itake 1y ago
      neat, so how does this mesh with OpenAI (and deepseek) offering country-specific models? Why is it ok for OpenAI to do this, but everyone is up in arms when their competitor does?
      - nerevarthelame 1y ago
        I don't know what regionalization OpenAI or Deepseek do. But it makes sense that they would change some things because of different languages, cultures, and regulations. Most global businesses tailor products for different regions. People are up in arms that Grok is using their CEO's shitposting as a primary knowledge base because that is a low quality source of information.
        itake 1y ago

by itake 1y ago

unshavedyak 1y ago

It is pretty interesting that this model will have two forms of bias though. One model derived from the company perspective and its training data, and two from Elon himself. Months ago this model would have promoted Trump, but now it'll call Trump disastrous for the economy. I don't know what to think of general company biases, and we've all been expecting biases to start favoring share holders eventually.. but biases based on twitter rants potentially changing day to day certainly is a new unique feature of Grok i guess.

czl 1y ago
> Months ago this model would have promoted Trump, but now it'll call Trump disastrous for the economy. There’s a well-known quote often attributed to economist John Maynard Keynes: “When the facts change, I change my mind. What do you do?”
- unshavedyak 1y ago
  Yea but this example wasn't of the facts changing. It was of opinions changing. Musk had one opinion months ago, and a different one now. The bot explicitly searched twitter for the current opinions of someone with a history of less than stable actions (i'm trying to be generous for sake of neutrality). This bot isn't at the whims of the corporate oversight or advertisements, or even direction of a CEO - the bot is at the whims of every post from a chronically online and addicted Twitter personality. Even if you ignore Elons other flaws, making a bot the sum total of everything he's said on Twitter is pretty impressive. .. and not in a good way.
  - czl 1y ago
    This isn’t just about opinions randomly changing. Musk’s views seem to have shifted in response to changing facts -- like the passage of a major spending bill, whichtie directly into his concerns about government debt. So it's not surprising that his stance evolved over time. The way the bot reflects his views is a bit awkward, I agree. I assume it’s something the Grok team will want to improve. One possible reason for the behavior is that if Twitter data was used during training, and Musk is a dominant voice there, the model may have learnedto rely heavily on his posts -- especially if that helped it predict or score better during training. To Musk’s credit,he’s pushed for more transparency. That at least lets people see these odd behaviors and raise questions. With other models, like OpenAI’s or Anthropic’s, it’s harder to tell when something similar might be happening. And while Musk definitely says someoff-the-wall things, his track record overall is hard to ignore-- whether one agrees with him or not.

spacechild1 1y ago

There is a slight difference between feature steering and intentionally installing the (de-facto) CEO as the principal source of truth.

itake 1y ago
Keep going. I thought Anthropic’s CEO is the source of truth that AI based on his belief that it should avoid these topics. Musk has different opinions than Dario, but they are both introducing biases into their respective companies
- nerevarthelame 1y ago
  Choosing not to answer - regardless of whether or not that was a rule mandated by the CEO (an unsourced and unlikely claim given the corporate structure of most large organizations) - is far different than insisting on an answer from whatever the CEO last decided to tweet. One is returning "null." The other is not. One says, "Figure that one out yourself." The other says, "Here is the truth."
  - itake 1y ago
    neat, so how does this mesh with OpenAI (and deepseek) offering country-specific models? Why is it ok for OpenAI to do this, but everyone is up in arms when their competitor does?
    - nerevarthelame 1y ago
      I don't know what regionalization OpenAI or Deepseek do. But it makes sense that they would change some things because of different languages, cultures, and regulations. Most global businesses tailor products for different regions. People are up in arms that Grok is using their CEO's shitposting as a primary knowledge base because that is a low quality source of information.
      - itake 1y ago