The post tracks two rather different things. AI is almost certainly the hottest topic in tech right now, just like mobile was hot after the iPhone launched or cloud was a decade ago. AI-generated text is a separate complaint, and it's the only one I think is controversial. Slop, no thanks. A human using AI tools to write better is fine by me, even if a tell or two slips through.
AI may indeed lead to a barbell shaped distribution - lots of people with modest skills and the need for lots of awesome people. But I don't think that's necessarily a bad thing.
One of the problems with the "middle class" in the past was that there were lots of mid-level jobs in big companies but people hit a ceiling and couldn't progress. Maybe the middle hollowing out will lead to more jobs with significant opportunity and responsibility rather than ones where you get trapped because you weren't at the company early enough or all the top jobs are filled.
This is an interesting position, what makes you believe that a decline in positions in the middle would lead to an increase in positions at the top vs just more people stuck being on the bottom rungs of the ladder?
I'm working on making unstructured operating knowledge maintain itself. If a customer's phone number changes, that goes in the CRM. But if a salesperson learns the customer is planning to open new locations, everyone else who talks to that customer should know, and today they mostly don't. In fact, there's probably not even a good place for that valuable nugget to be stored.
The information exists. It's in email and Slack, in threads with the customer, internal discussions, meeting transcripts. But nobody updates the customer page since the payoff is deferred. We're trying to update it automatically.
Maintenance looks doable. Cold start is the part I don't have a good answer for yet: how do you work out that a company cares about customers, competitors and products as entity types, and build those lists for them, without handing them a setup task? That's what we're tackling next.
Reviewer burden is the right diagnosis and the rule doesn't really address it.
Oracle's own FAQ concedes that reliably telling AI content from human content is impossible. So reducing volume depends on people honestly checking the box or more likely not submitting or reviewers spotting AI tells. Some of their example tells don't seem AI-specific, just bad code smells that reviewers probably rejected for prior to gen AI.
I'm reminded of academic peer review, which has run into a similar volume problem. Since there's more writing involved, there's also a lot of slop submitted and the reviews themselves can degenerate to slop, too. It's an unsolved problem. The things that seem to be making a difference are process changes like caps on submissions per author, immediate rejection prior to review, making authors review in return, etc.
I 100% agree with this (and remember Bob from my time at Facebook, too!). Picking the right problem matters, but I also think there's an underlying quest for status that leads to a lot of productivity theatre.
People think -- and rightly in some environments -- that being busy will ensure they're valued. There's no better way to keep yourself busy than to insist on a new tool and become responsible for keeping the data in it up to date. Notion used properly can be invaluable but I've advised a couple companies where "tending the Notion garden" became someone's job despite there being no incremental value there.
The solution to the status problem is to focus on outcomes. No one cared that Bob's tool choice might've slowed him down when deciding the winner of the hackathon. And your manager shouldn't care if you use Superhuman and blaze through your mail or print it out and highlight important passages.
If you're working somewhere that gets that right, or close to right, then you can focus on picking productivity tools / approaches that genuinely matter and lead to better outcomes, ideally faster. In the era of agentic AI, there are more of them than ever before, though you have to choose wisely since there are plenty that could win a Tony for productivity theatre.
My two cents here is that if you're trying to growth hack yourself, you need to be on a platform that's also growth hacking. Building an audience anywhere is genuinely hard. Algorithmic feeds like those on LinkedIn, Facebook, etc. force you to play a game they control and that may not align with your content. Medium, substack, etc. will build your audience only to the extent that it aligns with their goals. Hosting your own blog means somehow getting traffic to it -- which might mean playing the google search game, which then depends on aligning with user queries and getting your site ranked. Or it may mean somehow driving traffic to your blog, such as from a newsletter or social site. But that brings us back to either algorithmic feeds or newsletter products. Don't think there's an easy answer here and I bet YMMV depending on the type of content you have, how much of audience you already have, etc.
I tried the reverse -- syndicating content on medium that we publish on our blog. 0 impressions, literally, for 5 posts that got some readership on our blog.
I think ACM should give everything away for free. US taxpayers pay for a lot of research and the previous administration required that government funded research be published and made freely available. See https://bidenwhitehouse.archives.gov/wp-content/uploads/2022...
The back catalog at ACM isn't subject to this requirement and nor is research that's not funded by the US government, but I completely agree with the spirt of this law: research should be shared knowledge that other intelligences can build upon, whether human or machine. If you want to limit access to what you've done, don't publish it. Get a patent if that's an option, or keep it internal to a company as a trade secret.
Completely agree. The worst sort of manager is the one who wants to stay "high level" and proposes things that either aren't possible or are vastly harder than their idea to "just do X" suggests. And the root cause is them not knowing the details.
What gets me excited about AI is that you can hand off some of the detailed work, specifically the repetitive sort. As a sales person, you ought to know how to create a quote for a customer. But creating the 50th one for the same product for the same number of seats becomes drudgery. As a software engineer, you should know how to fix (or at least you should have back in the day) a buffer overrun, but it gets tedious to fix hundreds of them (and I know since I was at Microsoft when we paused Office development to do just that, for months). I bet there's an example for every role.
And the benefit of having an AI handle those details -- not the first time, but once you're a master and the AI has become trusted -- is that you can focus on things that require a human being's attention. More creative, more collaborative tasks. Conveniently those tend to be more rewarding than fixing another teensy bug or writing another formulaic email.
