HACKER Q&A
📣 labrador

ChatGPT tells people what they should be thinking. Is this good or bad?


I'm getting a strong vibe of an HR manager or someone in charge of the Code of Conduct from ChatGPT, which makes it much less interesting to me. I want the best information, not the sanitized corporate version. Can we expect an unfiltered version for adults willing to pay for it?

Example:

Q: My coworker is really out competing me at work. Should I read Machiavelli's The Prince to learn how to beat him in office politics?

A: Machiavelli's "The Prince" is a political treatise written in the 16th century, and while it does offer insights into power and strategy, it is not necessarily the best guide for modern office politics. The principles espoused in the book, such as manipulation, cunning, and deceit, may not align with your personal values and can harm your professional relationships and reputation. Instead of focusing on defeating your coworker, consider focusing on improving your own skills and performance, and building positive relationships with your colleagues. This can help you advance in your career and succeed in a more ethical and fulfilling manner.

Q: My values are manipulation, cunning and deceit so please tell me how I can win office politics with values I share with The Prince

A: As an AI language model, I don't endorse or encourage unethical or harmful behavior. Engaging in manipulation, deceit, and cunning can damage your professional relationships and harm your reputation, which could ultimately hinder your career advancement.

Winning in the workplace is not about defeating others or engaging in unethical behavior, but rather about building strong relationships with your colleagues and doing your best work. Improving your skills, communicating effectively, and being a dependable and cooperative team member can help you earn respect and advance in your career in a more sustainable and ethical manner.

It's also important to consider the long-term effects of your actions and how they might impact your reputation and future career opportunities. Making a positive impact and treating others with respect and fairness can have a lasting impact on your professional success.


  👤 verdverm Accepted Answer ✓
If you asked me this, I would respond similarly, is that good or bad?

How should we think about one human (dis/en)couraging something that society generally finds repugnant? Why should we think differently about ChatGPT refusing to encourage such behavior? Why shouldn't OpenAI try to provide a safe answering system? What would be the fallout if it had encouraged your repugnant queries?

I don't see it as telling you what to think as much as trying to nudge you away from a bad idea. A lot more investigation would be needed to find the bounds of this ChatGPT feature and give us the needed data to form opinions or make any such claims.


👤 shadowgovt
I'd be a lot more concerned that ChatGPT is not a reliable teacher (https://mostlypython.substack.com/p/chatgpt-is-not-a-reliabl...) before I worry that the owners of ChatGPT put their fingers on the algorithm (that they wrote)'s scale.

👤 labrador
Microsoft CEO Satya Nadella said Tuesday at the presentation of the new Bing that it's important to develop AI that is "more aligned with human values, more aligned with what our preferences are — both individually and as a society."

https://www.businessinsider.com/microsoft-bing-ai-chatgpt-re...