tu quoque or exemption to the rule?

This is an argument of mine recently. No one has accused me of this fallacy yet. It's something I was pondering after having said it:

 

If global warming is an issue regarding AI, then shouldn't it also stand to reason that using discord, not to mention many other online things that  also harm the environment? And so does that mean being against using AI while using all these other huge consumption of energy resources mean that one is  acting inconsistently with regard to what one is supposed to stand for? not that the their argument is invalid; it's not. But they are logically inconsistent with what they propose to value. and this is my own words. no AI lol  

The fallacy here is ignoring degree (or perhaps black and white thinking). AI uses considerably more power than typical online services to the point where many argue the spent resources are not worth it. For example, one may be against global warming but still be okay with eating Mexican, even though it makes them fart more. Yes, they are contributing to climate change, but the degree to which they are is insignificant compared to the benefit.

The argument you've presented seems to address a potential inconsistency in how people perceive the environmental impact of various technologies and activities. Let's break down the reasoning and address any logical fallacies, biases, or reasoning issues.

1. **Comparison of Impacts**: The argument implies that if one considers AI usage environmentally harmful, then logically, one should also consider the usage of other technologies (like Discord or other online activities) in the same light. This part of the argument is drawing a parallel between similar activities in terms of their environmental impact. The key here is the assumption that AI and these other activities are equivalent in their environmental impact, which might require further evidence or clarification.

2. **Inconsistency**: This is the crux of your argument. You suggest that it is inconsistent to oppose AI on environmental grounds while still engaging in other energy-consuming online activities. If someone claims to value reducing environmental impact, they should be consistent across all technologies and platforms with similar levels of impact. This is a valid point in terms of highlighting potential hypocrisy or selective attention bias, where individuals focus on a particular issue while ignoring others of similar importance or impact.

3. **Tu Quoque Fallacy**: The tu quoque fallacy involves discrediting an opponent's position by accusing them of hypocrisy without directly refuting their argument. While your argument touches on the inconsistency in behavior, you explicitly state that it does not invalidate the initial argument against AI usage on environmental grounds. By acknowledging that the original argument is not invalid, you are avoiding the tu quoque fallacy.

4. **Exemption to the Rule**: The argument might touch on this if it assumed some technologies are exempt from environmental scrutiny for no particular reason, but this isn't directly addressed here.

Overall, your argument raises a reasonable point about consistency. It does not fall into the tu quoque fallacy as long as the focus remains on discussing behavioral consistency rather than attempting to invalidate the initial argument solely on the basis of perceived hypocrisy. It's also important to have supporting information about the relative environmental impacts of different technologies to strengthen the argument further.

The issue I see with the argument you present is that it basically reduces activities down to two categories: Those that consume energy (and therefore produce greenhouse gases) and those that don't. That would likely make this an example of a false dilemma. In modern society, it's nearly impossible to live without using some amount of electricity and being responsible for the creation of some amount of greenhouse gases. So, the question isn't whether you're doing this, but how much.

When comparing the two things, I think the analysis could get kind of messy. You could calculate how much energy is used for a minute on Discord vs. an AI prompt. But, if you weren't on Discord, what what you reasonably be doing instead, and what are the energy costs associated with that? One of the ongoing conversations around global warming is the responsibility of the individual versus large entities like corporations. In the case of AI, there's an energy cost associated with the industrial scale rollout and training being done by the owners of the systems, and a separate energy cost associated with prompts being done by individuals. Similar to Discord, this would have to be compared to the energy cost of whatever processes AI is being used to replace.

You could probably make some assumptions, get some averages, and come up with an approximation of how the two things compare in terms of energy usage. Assuming that Discord does use less electricity than AI, you'd then have to determine whether it falls above or below the threshold of what would be considered acceptable, but that's entirely subjective.