>"He said that, based on talks with Uber's senior engineering leaders, he realized higher token usage did not translate into a proportional increase in useful consumer features."
Goodhart's law strikes again at someone with enough power to be both ignorant of it and make others suffer their ignorance. You cannot simply measure productivity by tokens spent just like you can't measure it by hours spent in a chair at a desk.
As soon as tokens stop stop being subsidized, heavy agentic use will become as least as expensive than paying an (entry level) employee. When this happens many companies will trade off havy tolen usage for (maybe a bit slower, bit less accurate) employees again.
It's amazing that it took months to figure this out. "Well we thought that if engineers are told to maximize costs through AI use, to consume as much as possible of a resource that costs us money, then obviously good things will happen. Imagine my surprise when it didn't turn out that way."
Imagine if engineers were ranked based on their AWS spend. People allocate VMs and fill databases with terabytes of random bits, to get to the top of the AWS leaderboard. If you don't do this, you're ranked at the bottom, and good luck at the next review cycle. Who could have expected that this is not the road to success?
Now we are going to get a new profession. Token Engineer! They will be experts on tokenmaxxing! The job growth that the billionaire CEOs promised us from AI is finally here!
What if... we stop for a moment, and then, after thinking for a moment, we stop hammering nails with a microscope, and stop using token usage as a metric of productivity?
You mean… increasing AI budget has no direct relationship with productivity and therefore revenue? It’s not that simple? But… my TEDx talk and LinkedIn ramblings…
> ... stop using token usage as a metric of productivity?
and tokenmaxxing is even worse due to https://en.wikipedia.org/wiki/Goodhart%27s_law because whatever you measure with tokens, once you start "tokenmaxxing" you have no measure to look at
If any company announces that they use token consumption as an employee performance signal, for me that's close to a red flag to stay away from that company.
No company with good engineering leadership should act like this is remotely a good idea.
As I have snarkily observed at work: if I go $100 over the meal allowance on my business trip, I'll have to have an unpleasant conversation with my manager or finance. If I use $500 in AI tokens unproductively I'll be recognized for being a top AI adopter.
I have seen this type of behavior happen many times in different companies.
For example, at more than one company I've worked for, if you wrote shitty code but got it into "testing" faster than anybody else, you are considered a superior programmer. And then, if you fixed the hundreds of bugs found in your code seen as an extraordinary programmer going above and beyond the call of duty.
The problem is that many companies which had reasonable leadership in the past with the advent of LLM AI started to make rushed (and dubious from my point of view) decisions - using token usage to evaluate an employee performance is just one of them.
Tokenmaxxing makes no sense, it is akin to write extremely inefficient SQL / Spark Jobs, full of cartesian joins, ultra skewed datasets, etc, just for the sake of using as much compute / memory / IO as possible.
This always happens when the metric becomes the goal, companies should nurture and foster an environment where AI is used in the most efficient way possible, first asking "do we really need an agent for this" and if so, what kind of agent is needed, what model, reasoning level, etc.
They should also promote projects that aim at saving tokens, increasing cache hits, codifying the information in ways such they use as less context as possible (graphs of knowledge are pretty good for this!)
I'd be interested to know if this is about individual employee AI usage, or use of AI tokens in production features, or both - and assuming both, what the split is.
I can see how Uber could burn unbelievable amounts of tokens if they start running internal features that run a bunch of prompts against every completed ride, or every customer profile, for example.
Or maybe this is about employee usage, but they introduced some stupid "you get evaluated on how many tokens you used" thing a couple of months ago when that was trendy and are just beginning to notice how much that cost?
Not all tokens are created equal. It's easy to use a ton of tokens by having agents work together in parallel. That's basically the equivalent as people spending time in meetings, hardly a productivity win. As with everything in development, results matter, how you get there doesn't (unless you're a bad manager).
many of these leading AI companies are operating at large losses and subsidizing users with VC money. Profitability will entail having to impose greater limits and raising prices, so this will reduce to some degree the value proposition of AI compared to humans.
It’s death for a tech company to be “done,” since that means no more growth. So they will all bloat indefinitely until they implode or get absorbed. It’s simply the fate of all VC-fueled startups.
The black bill that is coming that nobody is prepared for is that the value of a token varies greatly depending on the human. Companies will quickly find out its much better to give your top 10% engineers a lot more tokens and lay off your average engineers. The 10x engineer will become the 1000x engineer.
"He said that, based on talks with Uber's senior engineering leaders, he realized higher token usage did not translate into a proportional increase in useful consumer features."
He's saying that like it's some grand epiphany and not the most self-evident, obvious thing I've heard this month. Some of the literal dumbest people on earth are in charge of these major companies.
75 comments
[ 0.20 ms ] story [ 1557 ms ] threadGoodhart's law strikes again at someone with enough power to be both ignorant of it and make others suffer their ignorance. You cannot simply measure productivity by tokens spent just like you can't measure it by hours spent in a chair at a desk.
Imagine if engineers were ranked based on their AWS spend. People allocate VMs and fill databases with terabytes of random bits, to get to the top of the AWS leaderboard. If you don't do this, you're ranked at the bottom, and good luck at the next review cycle. Who could have expected that this is not the road to success?
I know it's sounds stupid, but what if
and tokenmaxxing is even worse due to https://en.wikipedia.org/wiki/Goodhart%27s_law because whatever you measure with tokens, once you start "tokenmaxxing" you have no measure to look at
No company with good engineering leadership should act like this is remotely a good idea.
For example, at more than one company I've worked for, if you wrote shitty code but got it into "testing" faster than anybody else, you are considered a superior programmer. And then, if you fixed the hundreds of bugs found in your code seen as an extraordinary programmer going above and beyond the call of duty.
Management is always measuring the wrong thing.
This always happens when the metric becomes the goal, companies should nurture and foster an environment where AI is used in the most efficient way possible, first asking "do we really need an agent for this" and if so, what kind of agent is needed, what model, reasoning level, etc.
They should also promote projects that aim at saving tokens, increasing cache hits, codifying the information in ways such they use as less context as possible (graphs of knowledge are pretty good for this!)
I can see how Uber could burn unbelievable amounts of tokens if they start running internal features that run a bunch of prompts against every completed ride, or every customer profile, for example.
Or maybe this is about employee usage, but they introduced some stupid "you get evaluated on how many tokens you used" thing a couple of months ago when that was trendy and are just beginning to notice how much that cost?
When will Uber (or your favourite company) be 'done'? They've been writing software for 16 years.
They match drivers to passengers. More software isn't going to increase the chance that I seek them out instead of taking a bus or train.
Will their software be finished in 20 years? 80?
Wrote about this and the impact of to jobs here: https://x.com/deepwhitman/status/2058324179506831372
He's saying that like it's some grand epiphany and not the most self-evident, obvious thing I've heard this month. Some of the literal dumbest people on earth are in charge of these major companies.