It looks like AI is not only not improving business and political leadership; it’s actually making it worse.
I’m sure you took notice of Sam Altman. the CEO of OpenAI, who has just been fired. As it turns out, he was one of Time Magazine’s Persons of the Year for 2023 that called him “perhaps the most powerful person in AI today,” in its list of the 100 most influential people in AI.
Hmm, I wonder whether Sam ever asked his AI chatbot adviser how he would sustain being the most powerful person in AI for the longer-term? I hope he didn’t so we can defend the apparently counterfactual notion that AI is really no more advanced than we mortals are in predicting the one and only path to true leadership.
I think there’s a rich mother lode of more dysfunctional leadership advice to be mined here. I just noticed – in today’s e-newspaper again – that Elon Musk’s X has just lost several of its top advertisers including Apple and Disney. Musk has also been one of the great promoters for AI. Same question: did he also ask his AI adviser(s) how he would maintain this position of influence?
Oh yes, Musk was also proclaimed Person of the Year by Time in 2021. Is there a pattern here??
Oh, yep there’s even more. Now for Sam Bankman-Fried, who, I am sure you already have surmised he was in Time Magazine’s Top 100 influential people in 2023 as well. Natch, he’s no longer CEO of FTX. But he has also predicted a bright future for AI too. I guess he didn’t consult FTX’s Oracle (if it had one), although we would suspect that it would have made no difference if he had.
Yes, I’m going to tackle another shibboleth here, namely Donald Trump. Why so? He just seems to fit. For starters (of course) he was Time Magazine’s Person of the Year for 2016, too!! He lost his position as Uber-CEO of the US in 2020, albeit neither he nor his followers have ever been to admit it.
And he’s trying again. Will the second or third time be the charm? What will AI do for him if he were to be elected? Can AI create a real leader out of the very raw material we see on hand there?
This is a really simple game. I am confident I could go on for ever to recall the names of people who have been awarded great honors only to be booted out shortly thereafter. It’s hardly an original thesis. Except this time round instead of linking the fall from grace to the public award we can link it to AI, which self-evidently failed.
In ye old days of leadership thinking (no listening needed) we would not have had AI to blame for what happened to the gentlemen above. We would just have talked about narcissism and how it is linked with toxic leadership.
There are various forms of toxic leadership, but they all generally lead to the same result; chaos and screwups. The forms include delusion, increasing arrogance and total unwillingness to listen to others. The biggie is being incapable of listening and taking advice from others. “Others” here includes AI. The quartet of CEOs I have pointed to above all are bad listeners.
The new forms of AI are all based on the so-called “large language models” (LLMs). They get their information basically by stealing it from others who actually supply the intelligence they provide via their writings which the AI companies get for free by scraping their websites.
This sets up a lose-lose relationship between CEOs and LLMs. The toxic leadership behavior CEOs don’t listen, so they can’t make progress to better leadership because even if and when it exists in the LLM advice, the CEOs are incapable of listening to and digesting it.
That means the CEOs and their leadership advisers aren’t producing the raw material insights about leadership which should constantly be in the process of being developed and sent over to the LLMs. So, the LLMs simply don’t get the materials they need to refine insightful LLMs for working stiffs.
So, what is rapidly becoming clear is that, so far, AI isn’t adding anything new to the party. CEOs that are believers so far don’t seem to be doing any better than those who don’t believe. Toxic leadership behaviors seem to be just as strong as they have ever been, maybe even stronger.
In fact, on the face of it seems that we should be backing toxic leadership behaviors as the best explanation of leadership failure as compared with anything that’s coming out of the apparently better next step of AI.
There are some interesting emerging theories of leadership such as behavioral economics and behavioral finance. These have pointed to psychological drivers they call cognitive biases as being a more granular explanation of leadership failure.
Ironically neither the vast majority of universities teach these ideas nor are the vast majority of CEOs aware of them, so these new insights are not making any difference. We can be pretty sure that AI isn’t telling any of its human mentees to figure out what their own cognitive biases are and how to correct them so that leadership teams everywhere can avoid even more severe existential global leadership crises.
Since ignorance is bliss, any emerging leadership ideas that are not being checked out and executed by the CEO clients of the AIs are just another channel for CEOs to refine the toxic behaviors that amplify even bigger screwups again. In other words, AI is going to reflect gaps in leadership knowledge by CEOs by ignoring behavioral economics and finance and cognitive biases not to mention other leadership insights that are yet to be discovered. In that case it won’t see any need to advise its mentees to lift their game.
In fact, it may well be that AI is actually reinforcing the lack of leadership insights in the market; it’s not seeing any of the truly promising ideas so it’s just ignoring them when it advises its leadership students on ways of approaching leadership for new days, ages and scenarios.
To that extent AI is itself a part of the problem of the coming new waves of leadership failures.
As long as that situation continues, toxic leadership behaviors will continue to be the overriding reason for leadership failures in both the public and private realms.
So, phooey to any transformational leadership insights from AI.
Remember, you heard it here.
When you subscribe to the blog, we will send you an e-mail when there are new updates on the site so you wouldn't miss them.