About

Frequently asked questions

Why don't you like AI?

We don't hate all AI - we believe that, like any tool, it can be used positively or negatively.

With that said, there are a lot of reasons we don't think it's appropriate in most contexts. You can find more information in our Info section. To summarise, we're worried about:

• The decrease in cognitive engagement and learning, and the increase in cognitive fatigue, that the use of generative AI seems to lead to.

• The impact of data centres on local communities, including electricity use, water use, land use, bushfire risk, light pollution, air pollution, noise pollution and rising cost of living.

• The impact of data centres on the broader environment, including on native species, waterways and temperatures.

• The non-consensual use of text, image, audio and video to train generative AI without compensation to the original creators.

• The threat that AI poses to certain job markets, especially entry-level jobs.

• The impact of generative AI on art, creatives and creative industries.

• The disparate impact of AI on diverse communities, including women, children, people in poverty, people of colour and First Nations peoples.

AI is also a very broad term, covering everything from chessbots to social media algorithms to large language models. We're mostly concerned with generative AI - AI which produces a novel output - although other kinds of AI have problems as well.

Do you use generative AI?

We can't speak for everyone in the group, but as for the founders: no, not at all. We do not find it necessary, and we have serious moral issues with how it is designed, resourced and used.

Do you think I'm a bad person for using AI?

No, definitely not. People can turn to AI for a variety of reasons, like not feeling that they have enough time, money, energy, or skills to do what they need to do.

Our problem is with companies, institutions, and governments that produce and use generative AI for inappropriate reasons and with insufficient safeguards.

How do I get involved?

Come along to one of our in-person events! We run events at UNSW multiple times a trimester, from open meetings to big get-togethers, so the best way to get involved is there. You can join our Discord or Whatsapp via QR codes or links at these events.

If you're unable to go to in-person events or want to get in contact with us quickly, the best way to do that is via the details on the Contact Us page.

Since we are a small group without the capacity for heavy moderation, we do not publicly publish links to join our Discord and Whatsapp groups.

Our Principles

Technology should be for the benefit of all humans.

The zeroeth law of robotics: “A robot may not harm humanity, or, by inaction, allow humanity to come to harm.” - Isaac Asimov

Many people see a bright future when they think of technological advancements. Writers have dreamed of utopias where we play instead of work, we have mastered the threats that face us, and all humans can live without existential stress or fear. Although we believe this is possible, it is naive to think that is the direction we are currently heading.

Governments, corporations, and the wealthy use technology to abuse and dominate the world’s most vulnerable populations. From social surveillance, to environmental harms, to modern slavery, the cruelty that technology is being used to enact is both unnecessary and avoidable.

Technology is not evil. Famously, Richard Feynman describes science as a key - itself neutral, but able to open the gates of heaven as well as the gates of hell. Technology, including AI, is just a key - it is our choice what to do with it. We must all commit to making the decision which best protects the interests of humanity.

AI requires robust safeguards and legal frameworks.

We have already seen the consequences of widespread AI. Environmental destruction, propaganda, surveillance, mass violence, sexual abuse, and the violation of labour rights are only a few of the many, many concerns. It is not enough to condemn it in words - we must have processes in place which prevent harms before they occur, and to rectify them once they happen.

Human rights need to be protected for everyone, regardless of their race, class, gender, ideology, or nationality. Individuals and groups need to have the ability to seek recourse against those that have harmed them, whatever those harms may be, and aggressors should not be able to use AI as an excuse. To protect the interests of everyone, robust laws and safeguards must be put in place.

When AI is used, the consequences should be seriously considered, justified, and appropriately mitigated.

Given that the potential consequences of AI use are so immense, it is obvious that using it without proper consideration is reckless at best.

The use of AI should be strictly avoided when human labour would produce a safer or better quality response, unless using human labour would be impossible from any practical viewpoint, or if it would pose too high a risk to the workers. If AI is to be used, all potential direct consequences (e.g. misinformation, misidentification, potential for misuse, physical harms) and indirect consequences (e.g. environmental and community harms from data centre use) should be evaluated in detail, and measured against the extent and likelihood of any potential benefits. If the potential harms have a substantial chance of outweighing the benefits, or if the harms are likely to occur to a population which may not receive the benefits, the project should be stopped.

When AI is used, continuous and robust monitoring should take place to prevent and mitigate potential harms. When a harm or potential harm is discovered, immediate action must be taken to mitigate it and prevent it from happening again.

Generative AI, in its current form, has no place in education.

After considering all available evidence, we conclude that the harms of generative AI far outweigh any potential benefits in education.

Generative AI wreaks havoc on local and global environments, invades privacy, undermines intellectual property, creates misinformation, relies on modern slavery, and has been shown to cause substantial psychological harm. Safeguards thus far have proven to be ineffective. Meanwhile, the research on potential benefits is mixed at best, with early evidence indicating that it creates dependence, increases fatigue, and atrophies skills.

Any educational institution that wishes for its qualifications to remain meaningful should discourage or ban the use of generative AI, except in niche fields of study where it is directly relevant.