Is ‘Around The Clock’ Too Much?
We all knew that the agentic explosion would change how we view work. But now, a couple of years in, we’re getting a clearer picture of what happens when you can suddenly delegate nearly all of the details of anything to a machine. Coding. Writing a white paper. Scheduling. Planning a timeline for a project. AI can do all of these things. So what happens?
The first larger trend we can see is exhaustion on the part of the humans who have to keep up with this stuff.
Issie Lapowsky of Bloomberg is out with a new piece on the burnout suffered by those who have to keep up with the hacking capacity of new models. A white hat named Val Mairet weighs in on the “AI madness” that recently supercharged the race between security pros and those bent on network destruction, aided by Mythos/Fable level assistance.
A study from McKinsey highlights these challenges, with this blurb:
“AI is … being used by bad actors as a tool to fuel more sophisticated cyberattacks and increase their volume, as exemplified by the rise in AI-enhanced social engineering and the substantial financial impact of data breaches. For example, gen AI has enhanced social-engineering techniques, in which attackers generate highly realistic phishing emails or deepfakes to trick employees into sharing sensitive information or credentials. In 2023, the total cost of cybercrime had more than doubled since 2015.”
So we can see the trends here: human operators, who need to sleep and eat and, eventually, live lives beyond their digital work stations, are being outpaced by AI agents that can work all the time.
In a very real way, the questions that are being asked lead to a broader philosophical question: what is the value of endless productivity? And it is a good thing or not?
“The question hanging over this entire transformation, of course, is whether all this building is creating more stuff of value, or merely creating more stuff,” Lapowsky writes . I also like this question that she posted in a forum discussing her work:
“Does the fact that AI agents *can* work around the clock necessarily mean they should?”
You can sort of sense that Lapowsky, here, is answering her own question . Churning out more stuff is not great if it can’t be aligned. Even if it can, to some extent, there’s issues with the inability to test, to do quality assurance. The company, as an implementer, gets a flood of product, but no testing.
And that’s only one aspect of what agentic AI may do, in transforming our lives. Just take a classical game theory approach.
I was looking beyond Lapowsky’s column, to other new headlines from Bloomberg, also from this reporter.
One is around the idea of AI as an “uninvited guest” to Zoom meetings. The idea is that a slew of AI agents can listen in on meetings, record things, make detailed transcriptions, and otherwise surveille the humans, without explicit buy-in.
“Some people view AI notetakers are a major time-saver that lets them be more present in meetings,” Lapowsky writes. “Others view them as a privacy hazard and yet another uninvited incursion of AI into their lives.”
Then there’s the demise of the summer internship, at least for the young people who used to benefit from them.
“An important rite of passage for college students is slipping away as companies use AI for more busywork and prepare for a future with fewer entry-level positions.”
Here again, let’s visualize this: you have a swarm of agents “coming out” to “volunteer” for the positions that the interns would have done. A surplus of hands, if you will, flipping the age-old saying “all hands on deck” into, well, a nightmare, to use Lapowsky’s term: the Sorcerer’s Apprentice, a tale as old as time.
When people talk about “productivity addiction” or the promise and the threat of “endless productivity,” this is what they mean. You have to ask: how much product can pile up in the corner, before you have a serious problem?
I’ll continue to report from the front lines of the AI boom, noting how people are responding, at MIT and elsewhere. I think we should all be pondering these questions hard, because they will have big ramifications for our lives.
Loading article...