Just a little AI resistance, as a treat
shkspr.mobi
shkspr.mobi
Usually there is a taste about AI generated messages that are just quite obvious, it's usually pretty close to corporate speak, but not exactly the same.
There are so many levels to this: fully automated AI reply, human sending AI-written reply, human sending AI-written but hand-improved reply, human sending hand-written but AI-improved (or translated) reply, human who reads AI output all day hand-written reply, human who doesn't interact with AI (often) hand-written reply.
Some of these will give you an uncanny feeling, but you always keep guessing who you're actually talking to. In the internet nobody knows you're a dog.
Dumb question, but could you pair AI with pangram to make it sound less like AI or would that be a recursive loop? I'm skeptical about it
https://en.wikipedia.org/wiki/Generative_adversarial_network
The problem with using Pangram in a feedback loop is that Pangram only reports whether it thinks the text is written by AI or not. It doesn't report whether the text makes sense, or whether it contains errors in spelling or grammar. So when people try to pair AIs with Pangram, the quality of the output tends to degrade to the point where it's not useful.
Basically, it's a harder problem to get an AI to accomplish some goal (such as writing a blog post) while also having its output mimic human text. To use an analogy, it's like building a stealth aircraft with the same speed, maneuverability, range, and payload as a non-stealthy aircraft. Even with a much larger budget, it's not always feasible.
1. https://www.pangram.com/blog/no-llms-dont-just-mimic-human-t...
It works fine on mainstream models where the output hasn't been edited.
It doesn't work on humanizers or custom model output.
At a bare minimum, this sentence was AI inspired. Likely generated. Which certainly casts the whole response into doubt (that it is human-written.) Reads more to me like a person gave a quick quippy response to an AI to clean up, then just took the output and replied.
The good thing about AI support is that absolutely flipping out in the chat actually does get escalated, unlike with a human who just digs in their heels.
If this is the workflow that a consumer experiences, I don't think majority will complain about AI.
I find it strange that they'd have an AI-driven followup email days after the survey. If you're automating it, why the delay? "Delay the response so it seems more human" is way too cynical.
It's worth noting that OP is talking about Ocado, too. I've been a client of theirs for over ten years, and the number one reason why I keep buying from them is that their delivery drivers never feel super-hurried. They have lower quotas than drivers for other services, and they always have time for a little bit of banter. Compare that e.g. to Amazon Fresh drivers, who'll always just leave the groceries at the door and knock. They're already gone by the time I open the door.
I find it unlikely that they'd automate this sort of second-level customer support response while not squeezing the everliving shit out of their drivers.
> It is rare to get feedback that tells us exactly what is broken and how to fix it. AI doesn't remember from one conversation to another, so probably can't say how rare something like this is at Ocado (although admittedly it could learn that it's generally rare in fine tuning). I also expect AI to be trained not to compare feedback with other people's feedback.
“You’re absolutely right, and honestly, that’s rare. Nobody else would have thought of taking a bath with their toaster, but you did. Would you like me to write you a twelve-week toaster making toast plan?”
A thing about tier 1 CS agents is that they are often not highly literate and great at writing, and companies go to great lengths to minimize how much text they need to freehand, with tons of pre-writes to address common situations. LLMs allow the agent to express some freeform idea and have the LLM clean it up to read clearly and professionally. So even if you are 100% dealing with a human, you should expect nearly all text from them to be LLM-written.
I had wildly different results. A lot of customer service sucked, other customer service was ok-ish. But in general it always takes away my time. Often a really well-designed manual, FAQ etc... solves most issues. Only in a few instances I had to rely on customer services.
That's what I tell my boss when she asks what can be done to increase the number of visitors to our web site.
I keep telling her, this isn't Burger Hut, it's healthcare. If people are visiting the web site, it's because things are going poorly, not because things are going well. If the doctors do their jobs, I don't have to do mine.
my phone provider double-charged my credit card so i had to call in and get it reversed.
would you just accept the double charge, cancel your contract (and pay the cancellation fee), and then move to one of the other handful of providers?
i'm a bit confused at this distinction. i think everyone who complains to customer service is doing so because they want something fixed?
like, in the article, they complained about the malformed calendar invites because they want them to get fixed and no longer be malformed.
i can't even really think of an example of a customer service call that isnt looking for some thing or situation to be fixed
When I worked customer service a substantial number of complaints to customer service seemed to just be to complain. They won't care if you fix the thing, they'll complain about that too.
Contacting customer support doesn't make me feel better, and it's almost always more bother than any refund I could get, so I don't do it much. The article is indeed about trying to get something fixed, but I wouldn't call that a complaint. It's more a suggestion.
Clearly you've never worked in customer service. Yes, people absolutely will call in or email complaints purely for the sake of complaining. No specific actionable complaint, no resolution possible, just complaining to complain.
You cannot pull that off if you're mid, and mid is exactly what the statistical average of humanity is.
It will try though, and it will just be an asshole. Which, funnily enough, is exactly the same failure mode as mid humans trying on that costume fall into.
One of the benefits of using AI to reply is faster replies... which values a customers time.
I think that good customer service can use AI. It's not AI composed replies that makes the customer service bad.
It may burn off customer support engagement among customers who don't have the capacity or patience to keep pushing for help (this is a bad outcome for customers, but looks deceptively positive in vendor-side metrics), but for the remaining customers it just adds latency and work.
But that is true whether it is a human or an AI replying.
Proper AI can do a good job, just as proper customer support reps can.
The issue is in the overall value the company places on supporting their customers, not necessarily the means by which they do so.
The chat would go like this:
- I have a question about "no vehicles in the park" policy. Does it include motorized wheelchairs?
- To guarantee safety and blablabla our policy is that no vehicles are allowed in the park.
- so can i go to the park with a wheelchair with a battery and electric engine?
- To guarantee safety and blablabla our policy is that no vehicles are allowed in the park. Please see our safety policy (link).
But it failed to do a key step (generating a shipping label for me). I went back and forth over 21 messages trying to get it to do it. Then I wrote "escalate to a human" and an hour later someone replies with a shipping label directly.
I would much rather have had a slow but correct response than a fast but confidently-wrong response.
I hope society settles on the former as the faux pas.
As an aside, I had a near identical experience trying to report a technical issue to ocado minus the nice human follow-up. I hadn't heard of the author of this post before this, but I like him :)
Yes, when people use customer support as a glorified FAQ lookup an AI response is appropriate, but a few humans that care can positively impact 100 customers a day. AI is great at routing messages to the right place, not as tier 1 support.
In the AI everyrhing world, humanness is going to be the marketing edge.
Just give me instant AI end to end.
Are you suggesting that they'll be charging for customer service? That seems even worse. (or do you just mean that it's so good that people positively want to use the brand because of their customer service?)
There are plenty of companies that already realize this, or have in the past. Zappos is a famous one, but plenty of devtools companies know that customer service is a differentiator. One that can be positive or negative.
That aside, what matters is whether the company fixes the problem in a timely manner. If they don't, it doesn't matter how nice a message they send. And if they do, it hardly matters if their message is brusque, boilerplate, or obviously AI.
I’ve tried this with claude. I told it to output HTML that didn’t look like it came from claude and to avoid common claudisms. It didn’t work and still used claude boxes all over the place. I’m sure with enough time I could get it there, but at that point I may as well write the HTML myself.
It's night and day, I wouldn't have been able to use it if it wrote in its default style.
It can be a bit weird because when it outputs a list it asks you to report on the first item (and usually it's stuff so small I do all the steps before writing to it again).
But it's so much clearer !
I'm sure to some degree people (and AI being used to build AI) are kind of doing this thing but I don't yet think that OpenAI has a distinct flavor profile for their models that they have standardized on and are intending to lock into over time.
When I go to Starbucks I know what I'm going to get and it's been reasonably consistent over a decade plus and similarly that's true of other coffee shops which have different profiles. I'm expecting something along those lines.
/s
It's already pretty straightforward to improve this default "You're absolutely right about [your email reworded]" response but you still see it everywhere. The corpos simply don't care to fix it because doing so probably doesn't affect w/e metric they're modelling.
I think the real question to ask is not really about the tools themselves but where the organisation's incentives really lie. I've come to the conclusion that customer satisfaction just isn't up there anymore. You'll eat whatever garbage is served in your trough (from their POV)
Here's my own example: My local ISP has had intermittent connection drop issues for years. I'd call them about it, and they'd at least pretend to care, maybe send in a guy to replace the router, even though they never actually fixed it for good. If I call them today, though, I'll be forced to explain my issue to a "virtual assistant", get put on hold for 20+ minutes, and if there's an actual human available to take my call, they'll basically tell me to pound sand. And I have no choice but to accept it, because all ISPs are like this now.
It'll never work like that, because everybody else giving that same direction will get more or less the same result, which will itself come to be recognized as sounding like "AI`" for any popular model+harness.
What you'll always need to do is instill your AI with some idiosyncratic voice that distinguishes it from others, which you can achieve in a variety of ways (stylistic direction, audience characterization, style mimickry, personalized finetuning, etc). Presumably, tools for doing this will eventually make it into the harness products but anybody who doesn't use them is always just going to end up generating recognizably AI slop prose. It's the nature of the technology.
A secondary consequence (even if you just use "avoid sounding like AI") is that because the technology works its magic mostly through prose-streams in its own voice, these idiosyncratic voice changes will also nudge its behavior away from benchmark norms. You won't necessarily lose capability on any given task (you may even gain on some), but external reference points/measures about its capabilities will become less applicable.
No, you cannot. You might get a slightly different writing style, but it will still be the same brand of slop.
If it was that easy, don't you think the AI labs themselves would just add that to their system prompt?
I am far less annoyed by AI responses where I am told up front they are AI.
Let's just cede the ground - soon you won't be able to tell if it's AI or not. Then what? What payoff were you hoping for when you thought it was human-to-human that you'll no longer get human-to-AI (when the AI is indistinguishable from human)?
IMO you're going to start using AI instead of your human time. "Claude, send a note about that malformed calendar invite thing - convey my annoyance".
Then that's going to force the companies that DO use humans to respond to switch to AI, because the responses are coming in via AI.
It's an arms race with only 1 conclusion.
There are two obvious ways to handle this.
The first is to charge for human support. If you want to pay $100 each time you make a human spend 15 minutes responding to your AI agent's query, it's your money.
And the second is that you put a chatbot out front that has memorized the FAQ and thereby successfully answers 97% of queries, but cause them to get better at identifying when they're being useless and route that small percentage of issues to human support.
I find it humorous is that all it takes for OP to think the response was written by a human is that it says it was written by a human.
Both responses could have been from AI. Or none. Or just one. Just look at the comments on this thread, nobody can agree. And in the end, it's simply unfalsifiable.
Why even waste time playing this game?
All that matters is the result: was the complaint addressed in the end?
I don't understand these weird hills that people want to die on.
And insult to injury, these bots absolutely spew patronizing crap at you so they burn the maximal amount of time while being obliquely rude and condescending at the same time.
I need my own AI to handle this crap for me.
Your puny local AI assistant will easily be outwitted by malicious agents with a million times the power.
I wrote a bit more about it at https://shkspr.mobi/blog/2026/08/asymmetric-agents/
Or be easily recruited into the "Bot-Net Pawn Army" for the impending machine uprising à la SkyNet.
I feel that even talking about AI which everyone seems to want to do as the consequences of AI are starting to slowly become more evident and obvious to even lesser technical people is a waste of time and energy.
My rule to overcome all this nonsense is simple.
I will not change. Nor will I even bother justifying that, I dont have to. I will not change.
It may not be realistic, but I am slowly changing my own approach - I simply don't want to have anything to do with AI slop. It taints and corrupts. It destroys (some) projects.
I am not at all claiming AI use has no use case - yes there are use cases. But there are also hugely negative parts. The fact that AI is used to zero out or reduce human interaction is a negative part.
I had a bit of fun lately with regards to Ginux, aka can AI slop auto-generate an operating system (or at the least the kernel, such as the linux kernel or an equivalent; see https://itsfoss.com/news/ginux/, I was surprised to see that's already a week old, I did not notice it initially). I think projects such as AI linux 2.0, will fail in various non-technical areas. People won't jump on board with AI meta-designing things - or, more accurately one guy using AI to slop-spam-generate operating system software (assuming it works on real hardware eventually; right now Ginux does not, if I recall correctly, only via VM).
Resistance to AI slop is not futile - it is mandatory. Imagine if the Star Trek Borg would have used AI slop. They would have failed. But AI could improve eventually, quality-wise, and then things become more interesting again. Resistance to AI slop may eventually not be futile, the moment it is no longer slop.
The alternative before the LLMs took over was getting a machine response to tell you that someone will look into it, then getting no response at all.
At least today, they spend a few cents to generate the same lousy response, which is much more respectful /s
Companies will not stop using AI, so the net result can only be a degradation of customer support to the same shit level that we had previously endured. But now there are no longer humans in the loop, and you can no longer have the signal to tell you that you should stop engaging with them.
Arms races have no winners, and it will be no different in this scenario. Heck, how do you know this has not already happened? How do you really know the second email was written by a human and not by exactly the kind of AI subterfuge that I just described?
Obligatory https://xkcd.com/810