Setting time on fire and the temptation of The Button

We used to consider writing an indication of time and effort spent on a task. That isn't true anymore.

I saw a bit more of the future of AI at work this week, and it shows every sign of vastly boosting productivity, while also causing a crisis of meaning in many organizations.

For such a dramatic statement, the actual bit of AI technology I got to experience this week is incredibly minor. It doesn’t do anything that AI couldn’t do before. In fact, other AI systems do it better. It is a button, added by Google to Google Docs. It says “Help me write.”

When you push the button, you are asked to prompt the AI and you get a draft of whatever document you asked for, courtesy of Google’s Bard. Simple. And something millions of people were already doing with ChatGPT and Bing and Claude (all of which are generally much better than Bard, today). And this same capability will be coming soon to Microsoft Office as well.

So why do I think this is a big deal? Because, when faced with the tyranny of the blank page, people are going to push The Button. It is so much easier to start with something than nothing. Students are going to use it start essays. Managers are going to use it to start emails, or reports, or documents. Teachers will use it when providing feedback. Scientists will use to write grants. And, just as we are seeing with Adobe incorporating AI into Photoshop, when AI gets integrated into a familiar tool, adoption become simple. Everyone is going to use The Button.

Now, there are a million implications to outsourcing our first drafts to AI. We know people anchor on the first idea they see, influencing their future work, so even drafts that are completely rewritten will be AI-tinged. People may not be as thoughtful about what they write, or the lack of effort may mean they don’t think through problems as deeply. We may not learn how to write as well. We may be flooded with low-quality content. All of these implications are significant, but I want to focus on one thing that, as an academic at a business school, really stands out to me: the coming crisis of meaning.

Setting time on fire

A lot of work is time consuming by design. Take, for example, the letter of recommendation. Professors are asked to write letters for students all the time, and a good letter takes a long time to write. You have to understand the student and the reason for the letter, decide how to phrase the letter to align with the job requirements and the student’s strengths, and more. The fact that it is time consuming is somewhat the point. The fact that a professor takes the time to write a good letter is a sign that they support the student’s application. We are setting our time on fire to signal to others that this letter is worth reading.

Or we can push The Button.

As you can see, from just pasting in the job description and a bare-bones description of the student: My name is professor Jo Lee, write a glowing letter of recommendation for Sally Doe, who got an A in my college entrepreneurship class. Make it for this job: Infosys Foundation [I pasted in the full description here], I got a letter in a couple of seconds.

Now, in Google Docs, The Button uses Google’s Bard AI, which remains pretty disappointing. You can see that the letter it produces is mediocre and includes a lot of extra made-up details that feel off. But, when we paste the same material into GPT-4 (which is being integrated into Microsoft Office), you get something pretty impressive. It is worth just skimming the text of the letter (produced from a single prompt!) to see what I mean:

Dear Hiring Committee,

I am writing this letter to enthusiastically recommend Sally Doe for the internship position at the Infosys Foundation USA. As a professor of Entrepreneurship at our institution, I had the pleasure of having Sally in my class last semester, where she consistently displayed her incredible intellect, strategic thinking, and dedication to her work.

In the classroom, Sally was a standout student, not only for the caliber of her work but also for her commitment and collaborative spirit. She earned an impressive 'A' grade in my challenging course, which is testament to her commitment and her capacity to comprehend, analyze, and execute complex concepts and assignments.

One of the remarkable traits that Sally exhibited in my class was her knack for digital learning platforms, which makes her an ideal fit for contributing to the development and management of the Pathfinder's Online Institute. She showed an excellent grasp of modern technologies and was often at the forefront of discussions on enhancing user experience and design.

Additionally, Sally's excellent communication skills, both written and verbal, set her apart from her peers. Her assignments were always articulated in a clear and engaging manner, revealing her ability to draft compelling narratives. I am confident that she will utilize this skill to significantly contribute to your social media and marketing efforts.

Furthermore, her ability in managing and analyzing data would be invaluable in supporting grants management and the development of an impact report. Sally's analytical acuity, attention to detail, and proficiency in extrapolating qualitative and quantitative insights from complex data sets were seen in several class projects.

In terms of events, Sally played a pivotal role in organizing and implementing many of our college events and campaigns. Her leadership, creativity, and ability to work under pressure have always been applauded by her peers and faculty alike. These experiences and her skills would be extremely beneficial for the planning and implementation of the Pathfinders Summer Institute.

What sets Sally apart is not just her academic and extracurricular accomplishments but also her passion for education, particularly in the field of computer science. I have seen her actively researching and engaging in discussions about K-12 education trends. Her interest in this field will undoubtedly bring fresh insights and drive to your team.

In conclusion, I have no reservations in giving my highest recommendation for Sally Doe. I firmly believe that she would be an exceptional addition to the Infosys Foundation USA and contribute significantly to your various projects. She exhibits the qualities necessary for success in this role: initiative, intelligence, and a highly cooperative nature.

Thank you for considering my recommendation. Should you require any further information, I would be happy to provide it.

Best regards,
Professor Jo Lee

This is a good letter of recommendation, responding point-by-point to the details of the job, and suggesting a real knowledge of “Sally” (who is entirely fake). You will also notice that it makes stuff up, but in a way that is pretty plausible. It would be easy to make the letter more correct, either by providing more material in the prompt or by interacting with the AI: Incorporate the following real examples: ____. Replace the story about her extracurriculars with praise for how she did on this assignment: _____. And so on. Someone experienced with ChatGPT and Sally could make this letter factually correct in a few minutes, rather than spending ten times as much time writing a letter from scratch.

And the terrible, horrible thing about it is THIS IS A GOOD LETTER. It is better than most letters of recommendation that I receive. This means that not only is the quality of the letter no longer a signal of the professor’s interest, but also that you may actually be hurting people by not writing a letter of recommendation by AI, especially if you are not a particularly strong writer. So people now have to consider that the goal of the letter (getting a student a job) is in contrast with the morally-correct method of accomplishing the goal (the professor spending a lot of time writing the letter). I am still doing all my letters the old-fashioned way, but I wonder whether that will ultimately do my student’s a disservice.

Now consider all the other tasks where the final written output is important because it is a signal of the time spent on the task, and the thoughtfulness that went into it. Performance reviews. Strategic memos. College essays. Grant applications. Speeches. Comments on papers. And so much more.

With everyone pushing The Button for most emails, documents, and even (soon!) spreadsheets and presentations, what documents mean is going to change fundamentally, and that is going to spill over to our work.

The crisis of meaning

We care about the meaning of our work, and generally feel like we are doing something valuable. It may be surprising, but the vast, vast majority of people (92%!) think their job is socially useful. Even in those professions where workers are most cynical about their job (food prep workers, economists), nearly 80% or more still consider their job to be useful. And, indeed, though every job has meaningless tasks (especially at big organizations!), we mostly take pride in what we do.

So, at first, the ability to outsource crappy, meaningless tasks to the AI is going to tremendously freeing. The worst parts of your job go to AI so that you get to focus on the good stuff. That is indeed what we have seen in some early research on using AI for work. People who use AI enjoy work more, and feel that they are better able to use their talents and abilities.

Change in job satisfaction for people assigned to use AI (treated) versus not (control) on writing tasks, from Noy and Zhang

But then The Button starts to tempt everyone. Work that was boring to do, but meaningful when completed by humans (like performance reviews) becomes easy to outsource - and the apparent quality actually increases. We start to create documents mostly with AI that get sent to AI-powered inboxes where the recipients respond mostly with AI. Even worse, we still create the reports by hand, but realize that no human is actually reading them. This kind of meaningless task, what organizational theorists have called mere ceremony, has always been with us. But AI will make a lot of previously useful tasks meaningless. It it will also remove the facade that previously disguised meaningless tasks. We may not have always known if our work mattered in the bigger picture, but in most organizations, the people in your part of the organizational structure felt that it did. With AI-generated work sent to other AIs to assess, that sense of meaning disappears.

I don’t see a way that we avoid dealing with this coming storm. So many important processes assume that the amount and quality of our written output is a useful measure of thoughtfulness, effort, and time. If we don’t start thinking about this shift in meaning now (which does not require any AI more advanced that what is currently available), we are likely to face a series of crises of meaning, as centuries-old approaches to filtering and signalling lose all of their value. It is already happening throughout education, as essays are increasingly AI-generated. It is going to happen elsewhere, soon, as everyone gets access to their own version of The Button.

But there is a potentially empowering outcome as well, if we want to pursue it. We can view the destruction of busy-work as freeing. We do not have to set our time on fire as a signal. We don’t have to do work that is mere ceremony. Doing that will require thoughtfulness from organizations and leaders as they redesign work for our new AI-haunted world. But the incentives are huge as well. Stripping away meaningless work removes a huge burden from workers, while reducing inefficiencies and broken processes. This is an amazing opportunity, but only if we are forward-thinking about the future of a world where most work starts by pressing The Button.

We can redesign work to free us from the Boschian nightmare of meaningless tasks.