Employees grow tired of constantly having to double-check and fix their AI-obsessed coworker's 100% AI-generated work output: 'It's low effort, low quality, and I don't want to be associated with delivering [low-quality] work as a team'

You'll find coworkers using it frequently who, undoubtedly productive, are shoveling AI-generated work out faster than anyone can keep up with, but when it falls to others on the team to QA or verify the AI output's accuracy and authenticity, then you are unfairly just pushing your own work onto others.

Where should employers draw the line between encouraging AI adoption and requiring employees to demonstrate they actually understand and can audit and verify the work they're producing?

Especially in workplaces where leaders have adopted an “AI optimistic” direction, flagging that a coworker is using AI will only get you branded as “anti-AI.” So, the only thing that can be done, as others have suggested, is to call out documentable instances where the work is not up to quality and let the team member whose work is in question explain themselves about how they generated it with zero editing, auditing, oversight, or quality control measures of their own doing.

Keep reading to see more of the source story and my thoughts on the subject. 

AI has created a divide in the workforce. There are strong supporters or decriers on every side. 

There are those who fervently swear that it has made them more productive or added something somewhere to their workflow that they were otherwise missing. If you challenge their belief, they will readily tell you that being anti-AI is the exact same thing as those who were anti-internet as it came into prominence at the end of the last century. 

There are those who stand on principle and refuse to even open an AI chat in their browser, keenly aware of the budding social and environmental consequences of the technology.

There are those somewhere in the middle who aren't quite decided how they feel about where the new tech will lead us, but they are surely wondering if the email or Slack message they just generated with genAI was worth the gallon of fresh drinking water and draw on already-strained electrical infrastructure it took to do so. 

Meanwhile, the Tech billionaires pushing the technology can't give a straight answer about what good the technology will serve for the general bulk of the workforce in the long run. And have, suspiciously, a seemingly persistent ability to let the not-so-occasional Freudian slip slide through about how there is no other intention other than the replacement of thousands of jobs. 

The speed at which menial tasks can be handled is appealing, and it can kind of get you in the right place. But you really can't help but wonder if the temptation to turn to it for smaller uses, and its relative success at handling problems in those areas, is just misleading its usefulness for solving larger problems. And by the time you've attempted to use it for larger projects and dealt with increasingly limiting context windows, where the earliest information slides off the back end of your project into seeming nothingness without any sort of indication that it is doing so, you find yourself repeatedly wishing that you had just done the hard yards of the smaller tasks yourself so that you had a better grasp on the project.

And even without context windows being reached, there is the issue of accuracy for critically important work and the security of sensitive information. Information fed into a private organization's database is by no means private or secure, and by the time you've seen an AI confidently give you the wrong answer to something, you're going to be left questioning everything else that it tells you.

Besides, once you have worked with AI-generated content much at all, you quickly start to recognize the signs of AI-generated content. And it won't be long before any text written, app coded, or artwork generated with AI is easily recognized by the population at scale.