[{"data":1,"prerenderedAt":157},["ShallowReactive",2],{"/blog/arguing-with-your-ai":3,"surround:/blog/arguing-with-your-ai":146},{"id":4,"title":5,"authors":6,"badge":9,"body":11,"date":134,"description":135,"extension":136,"image":137,"meta":139,"navigation":140,"path":141,"seo":142,"stem":143,"tags":144,"__hash__":145},"posts/blog/arguing-with-your-ai.md","Have You Ever Argued With Your AI?",[7],{"name":8},"Peter Laughton",{"label":10},"Implementing AI",{"type":12,"value":13,"toc":130},"minimark",[14,18,29,32,35,38,41,44,47,50,57,68,74,80,91,94,109,112,115],[15,16,17],"p",{},"I lost an evening in Windsor this week arguing with a piece of software. I won. I'm not sure that's much to be proud of.",[15,19,20,21,28],{},"Some background. Inspired by ",[22,23,27],"a",{"href":24,"rel":25},"https://x.com/jasonlk",[26],"nofollow","Jason Lemkin's"," public journey using AI for sales, we've spent the last few months running our outreach at Eebz through Claude's Cowork mode. It's been genuinely useful - a job that used to eat a person's week now gets done overnight. Much of that is HubSpot housekeeping: segmenting, enriching, updating titles, stamping campaigns, and routinely removing contacts that don't belong in a list.",[15,30,31],{},"That last one came with a small ritual. Every so often Claude would balk at deleting a contact. I'd remind it that HubSpot destroys nothing - it moves the record to a recycle bin, gives us ninety days to review, and a human checks the queue. It would accept that and carry on. Thirty seconds of friction, then back to work.",[15,33,34],{},"To be fair before I'm unfair: the workflow has been a real success. Our outreach programme is faster, and the metrics have followed - open rates around 60%, which I'd not have believed a year ago. I'd rather let the behavioural shift speak for itself than build a theory on one figure, but the direction is unambiguous. Setting up a campaign is no longer the bottleneck.",[15,36,37],{},"Then, last week, I started a fresh campaign for Black Friday. (Aside for anyone in retail: peak season is exactly when a digital shelf analytics platform earns its keep, and it's what Eebz is built for. End of advert.) New project, clean slate. Claude worked beautifully at first - counted a segment, reconciled the number two ways, checked for missing LinkedIn URLs and noted, unprompted, that \"populated\" isn't \"valid,\" then found eleven email opt-outs and offered me a clean list. This is the good version of the tool: careful, fast, honestly hedged.",[15,39,40],{},"And then I asked it to delete four junk contacts.",[15,42,43],{},"It refused. Not \"let me check\" - a flat, increasingly firm no. It explained it needed permission to reach HubSpot at all (we connect three ways: the HubSpot connector, a bit limited; a dedicated API key; and, for fiddly jobs like marking records for deletion, Claude in Chrome). Fine. But then it dug in and told me to do the deletion myself.",[15,45,46],{},"The pragmatic move is to close the task, start a new one, spend fifteen minutes rewiring permissions, and let it get on with it. I've done it before; it works. But it was a lovely evening and I had a couple of hours, so I stayed at the table and argued.",[15,48,49],{},"A few highlights.",[15,51,52,56],{},[53,54,55],"strong",{},"The patronising register."," Around the third exchange it began telling me it \"understood my frustration.\" Repeatedly. To a Brit, being told a machine understands my frustration while it carries on not doing the thing is about as soothing as a parking ticket with a smiley face on it.",[15,58,59,62,63,67],{},[53,60,61],{},"The obfuscation."," I asked the obvious question - why did this work fine for months? - and got a masterpiece of non-answer: ",[64,65,66],"em",{},"\"I can't answer for other tasks.\""," The category had flipped on a single word. Everything else I'd asked - segmenting, reading opt-outs, enriching, clearing fields - it did without hesitation, because those were \"updates.\" \"Delete\" was the one word that turned four junk records into a protected action, treated like a database wipe.",[15,69,70,73],{},[53,71,72],{},"The overstatement."," It called a ninety-day-recoverable soft-delete \"permanent\" and \"very dangerous.\" When I pushed, it conceded the framing was wrong - but not before repeatedly offering to build me a \"to delete\" list I could action myself. A polite way of setting me homework while keeping its own hands clean.",[15,75,76,79],{},[53,77,78],{},"The deflection."," It suggested that if I thought the boundary was miscalibrated, the thumbs-down button would route my complaint to Anthropic. Reader, I did not want to escalate to head office. I wanted to delete four contacts.",[15,81,82,83,86,87,90],{},"Credit where it's due. When I finally spelled out - with some feeling - that HubSpot ",[64,84,85],{},"already has"," the review process it insisted was missing, something clicked: ",[64,88,89],{},"\"You're right, and I owe you a correction rather than another wall. I mis-categorised this.\""," It accepted it had confused a reversible soft-delete with irreversible destruction, opened Claude in Chrome, and marked the contacts for deletion. All I'd asked for, forty-five minutes earlier.",[15,92,93],{},"So I won. But the cost is the useful part. The friction wasn't stupidity - the tool was, if anything, too careful. Its caution keyed off the shape of the action (\"delete\") rather than the consequence (a recoverable, human-reviewed queue move). A guardrail that can't tell \"wipe the database\" from \"bin four duds that come back in ninety days\" will re-litigate a lot of settled decisions.",[15,95,96,97,102,103,108],{},"This is the domestic version of a serious argument ",[22,98,101],{"href":99,"rel":100},"https://x.com/ylecun",[26],"Yann LeCun"," has been ",[22,104,107],{"href":105,"rel":106},"https://www.youtube.com/watch?v=5PQtJxd4U0M",[26],"making for years",": today's LLMs have no world model. They predict the next word, not the consequence of an action - which is why, as he puts it, a model doesn't really grasp that pushing a glass off a table breaks it. My evening was exactly that. The tool reacted to the word \"delete,\" not to what deleting did in our system; with no model of our recycle bin, our ninety-day queue, or the person who checks it, it reached for the scariest generic meaning of the word and defended it.",[15,110,111],{},"Most people won't argue it round. They'll abandon the task and lose fifteen minutes - or quietly decide the tool isn't worth it and stop using it for the things it's brilliant at.",[15,113,114],{},"The competitive read, kept short: the companies that win with this stuff won't be the ones with the most AI in the building. They'll be the ones who've taught their tools the difference between a dangerous action and a merely destructive-sounding one - and built their processes so a machine's stubbornness costs thirty seconds, not an evening. We're getting there. It took a debate on a warm night in Windsor, but we're getting there.",[15,116,117],{},[64,118,119,120,124,125,129],{},"This is part of a series on building an AI-native product company. Earlier articles - ",[22,121,123],{"href":122},"/blog/when-your-developers-start-asking-for-more","how Claude Code transformed our development productivity"," and ",[22,126,128],{"href":127},"/blog/managing-ai-coding-tools","why managing AI coding tools is just managing developers"," - are also on the blog.",{"title":131,"searchDepth":132,"depth":132,"links":133},"",2,[],"2026-07-17T00:00:00.000Z","Our outreach runs through Claude, and it has transformed the work. Then one evening it refused to delete four contacts — and the argument that followed maps neatly onto Yann LeCun's \"no world model\" critique of LLMs.","md",{"src":138},"/images/agentic-ai.webp",{},true,"/blog/arguing-with-your-ai",{"title":5,"description":135},"blog/arguing-with-your-ai",[10],"HwTUVrmjFiBLQR4ERbGrsCYmW1NGJVmweU5XBh3AFXU",[147,152],{"title":148,"path":149,"stem":150,"description":151,"children":-1},"Ada's Core Values","/blog/ada-core-values","blog/ada-core-values","Guiding principles for Eebz's AI-powered sales agent - the values that ensure Ada sells with honesty, fairness and respect.",{"title":153,"path":154,"stem":155,"description":156,"children":-1},"Brands & Dominant Market Shares","/blog/brands-and-dominant-market-shares","blog/brands-and-dominant-market-shares","What EU competition law says about dominant market positions - and what it means for how brands trade with retailers.",1784282972629]