Ways to buffer AI’S negative impacts
As AI 'happens', how can we protect ourselves from its negative and potentially dangerous impact our psycho-social health?
The artificial intelligence horse bolted a long time ago. Now, with the global political ‘vibe-shift’ to the right, the AI arms race is on, and conversations have shifted away from the risks and implications of AI’s hurried journey.
Meanwhile, AI's psycho-social impacts on humankind’s mental health have yet to be adequately examined beyond a few trending podcasts by Silicon Valley leaders and a few journal articles written by psychologists that would take effort to find, read, and digest.
In time, computer systems will complete most tasks that human intelligence normally requires. This sounds simple and obvious, but it will change humanity as we know it.
AI is more than a tool. It’s not like the invention of the radio or TV. It will rewire our societies and likely our brains, too. Consequently, it’s up to humans to cultivate a way of being and get ready to buffer its potentially damaging effects.
There are many risks related to its safety and mental health. Yes, there may be infinitely positive benefits—particularly if you’re looking at it from a technocratic viewpoint (characterised by the government or control of society or industry by an elite technical expert).
Either way, it’ll be a more than a bumpy ride.
We can try to adapt and internalise AI’s impact, but will our psyche and soul be okay? If we try, I guess it will require deep work, support from mental health experts, connection to the community and intentional choices to buffer what’s about to unfold.
Even if we assume that mass global disasters don’t unfold, such as weapons automation or an uncontrollable self-aware AI–even its most ‘diplomatic’ use could be harmful.
As jobs are cut, businesses automate, and prospects such as the universal wage are debated — can we both learn how to augment AI tools/agents for our purposes and cultivate ways to protect ourselves from them? We don’t know.
However, here are some of my best guesses of how individuals could prepare.
Do homework on the exponential benefits of AI.
This isn’t about P-Doom (the probability that the rise of artificial intelligence will have a disastrous effect on humanity) or ‘grokking the future’ (a reductive, often optimistic belief in massive leaps in machine learning). This is about balance so that we can define the actual risk.
To become cognitively aware of AI’s true power, we need to examine the potential benefits proclaimed experts are currently debating to ensure safe conversations about the risks.
As Anthropic’s C.E.O. Dario Amodei said in “Hard Fork”, The New York Times podcast about 2025 Tech Predictions and Resolutions (not verbatim):
‘If we don’t understand its benefits, it’s hard to understand the actual risk… which is a very dangerous thing’.
So before our governments and social media feeds polarise our views, let’s do the homework and look (for ourselves) at these benefits and why the world is even doing this.
Using consumer-based AI products (as they are now) can lead to denialism.
We’ve all opened an AI app and considered how useless it is for higher-order reasoning, thought, complexity, creativity, and the ability to plan. However, this will rapidly change from reading AI leaders’ position papers. You’ll discover that AI's power is exponential despite how you may feel now. While there are valid arguments that text-based machine learning is forever limited as it isn’t based on gaining information visually (like a human might in the physical world), we are entering the AI digital labour age. While worth reading, government safeguards and governance guidelines become outdated faster than you scroll.
Track how fast AI is adapting and transforming.
You may find this helpful, or you may feel overwhelmed by it. But it will give you a deeper understanding when pontificating with colleagues about it at your work’s water cooler. From me, I know there’s a lot more I could understand. However, there is a universal sense that AI is a threat, and generally, current regulators aren’t grasping what it all means. So listen to debates, test AI products, engage in the topic behind the algorithm set on your Instagram, and think more deeply about what it’s doing today and what it might do tomorrow.
Prepare to manage shock, loss, and grief related to your job and identity.
We already know that traditional job roles can become easily automated. Meanwhile, other more complex jobs, from coding to content creation, will likely be transformed, and often, positions will be reduced to those who can manage AI and make the best decisions. In preparing for losses, possible negative impacts may work hand-in-hand with the broader economy. For example, AI could be weaponised during an economic recession, particularly against populations deprived of access to socially and economically valued roles.
Applying a critical lens to fake content will become almost impossible: Understand the power of images and videos.
Who cares about deepfakes when we know the AI-generated content is misleading? We can ignore it! False. It’s been scientifically proven that this stuff ‘imprints’ on our psyche, even if it’s not real.
It seeps into our working memory, and we are impacted by it, which creates a cognitive load and emotional stress that we don’t need. The fear it produces can stay with us for some time. We will be overloaded with it, which explains why we must find ways to buffer its impact.
Things get much worse when we can’t tell if the AI-generated content is real. There may be so much fake information that people’s psyches get further distorted, creating junk land for us to filter through.
Commit to ‘hard-won’ skills/hobbies.
AI is going to make everything easier. We already know the downside of convenience culture. Well, this is convenient for the mind, too. Even the journey of a typically challenging mathematical or creative task will be reduced to ease.
Notably, the accuracy will depend on the quality of your AI service because more money equals better responses. However, it will remove the ‘hard-won’ aspect of many daily tasks, eliminating the possibility of dopamine being released after working long and hard at something. In many cases, the discovery/gestation process will be annulled.
Further, I would add that committing to ‘hard-won' skills without expecting financial reward might be an even more powerful practice. As humans complete the tasks they usually complete for money, is there still an intrinsic sense of meaning and purpose?
This may be an experiment to have with yourself. Can I do what I do for a job for free? Will it give me meaning?
Here’s a checklist that includes the obvious stuff to help buffer AI’s negative effects on your mental health. Many privileged AI working bees in Silicon Valley firmly but casually speak to all of these;
Get to know the potential benefits of AI so you can better understand the risks
Recognise the potential loss & shock it may bring; jobs, security, equitable systems
Investigate new career pathways with slightly better odds (many AI experts struggle to name what those may be)
Upskill in AI, apply it to your current industry, test/learn, share learnings
….While also focusing on life beyond machines:
Practice de-escalation from doom-scrolling, screen saturation, and AI use
Commit to ‘hard-won’ work or hobbies, even without fiscal reward
Honour and support local craftsmanship & the wisdom from traditional creative trades
Go outside, exercise, sleep, repeat