I'm done coding with AI
Credibility score: 53/100 β Mixed Credibility. Several questionable claims detected. Watch with healthy skepticism.
BSmeter analyzed "I'm done coding with AI" and rated it 53/100 for credibility (a BS score of 47/100 β mixed credibility), on 2026-10-16. Its weakest claim β "Claims AI has a 'negative impact' on the world β Loaded Language" β scored 45/100 and was flagged as loaded language. 30 claims were checked against the video transcript. Scores are produced by BSmeter's AI analysis of the transcript, not independent human verification.
Of 30 claims analyzed: 0 scored under 40, 25 between 40 and 69, and 5 at 70 or above.
Claims analyzed
Claims 20 years of expertise and a massive shift due to AI β No Frame (75/100)
At 0:03
Setting the stage with actual experienceβno obvious tricks here π‘οΈ
Why this score: He's establishing his authority through tenure before making a pivot. It's a straight setup to justify why his opinion matters.
Original quote: βI'm a software developer who's been coding for over 20 years. Over the last 18 months, my career, my craft, my hobby has been drastically changed by the introduction of AI coding toolsβ
Declares a total lifestyle pivot away from AI technology β Personal Story (60/100)
At 0:22
A massive life pivot based on a personal realization. Bold move π
Why this score: This isn't a scientific claim about the efficacy of AI, but his personal decision to abandon it entirely. It's a subjective stance on how he operates.
Original quote: βI have decided to stop using those tools and those technologies for coding and for anything else in myt life.β
Claims AI has a 'negative impact' on the world β Loaded Language β Loaded Language (45/100)
At 0:39
Using heavy-handed terms like 'negative impact' without defining the scope. It's a vibe, not a metric π
Why this score: He jumps straight from personal feelings to global consequences. He's using high-stakes emotional language to frame a broad, unquantified societal problem.
Original quote: βfor over a year and reflecting on how it made me feel, what the resulting code was like, and learning about the actual negative impact that these technologies have on the world at large.β
Sources: The bad and the ugly: AI is harmful, unreliable, and running out of data, People really hate AI, so why canβt they get enough?, AI is making everything worse
Recalling GitHub Copilot launch β No Frame β No Frame (75/100)
At 2:17
Just stating the facts about Copilot's function. No tricks here, just a straight observation π‘οΈ
Why this score: He's describing the actual functionality of a known tool without trying to sell it or exaggerate its impact. It's just the baseline reality.
Original quote: βThere was GitHub at some point, I think, launched Copilot in the Visual Studio Code editor. And it would auto-complete what you were typing.β
AI autocomplete is a distraction, presented as personal experience β Personal Story (60/100)
At 2:36
Just a guy sharing his vibe on autocomplete. Not exactly data-driven, but it's his reality π€·ββοΈ
Why this score: He is describing a subjective feeling of 'distraction' rather than measuring cognitive load or productivity loss. It's an anecdote, not a scientific study.
Original quote: βSo, if you had a function like sort items and you started to type sort items, it might show you 10 or 12 lines of what that function could be based on the information it had about what you were typing. I tried this and found it very distracting.β
AI hallucinated a non-existent software version 4.6 β No Frame (75/100)
At 3:26
A specific, verifiable disaster. He's not selling a lie; he's reporting a glitch π οΈ
Why this score: He provides a specific detailβversion 4.6βthat can be checked against actual software libraries. It's a concrete example of AI hallucination.
Original quote: βthe AI chatbot had made up a version of a key software dependency that didn't exist and was claiming that upgrading the dependency to version 4.6 would fix the problem and give us what we needed. But, version 4.6 didn't exist.β
A product lead was a 'true believer' in AI β Loaded Language (45/100)
At 4:19
'Absolute true believer' is heavy-handed. That's a character study, not a job description π
Why this score: He uses loaded, almost religious-sounding language ('true believer') to describe a professional's stance, which colors the reader's perception of that person's competence.
Original quote: βA person was hired as the product lead who was an absolute true believer in AI.β
Sources: From skeptic to true believer: How OpenClaw changed my life | Claire Vo, AI influence in product leadership: How teams must adapt, How The Leading AI Companies Do Product (Part 1) β Ant Murphy
Claims AI is often wrong and no faster than him β No Frame (75/100)
At 5:48
Just a straight-up reality check on the current state of tools π οΈ
Why this score: He's not selling a dream; he's calling out the friction and error rates that developers actually deal with daily.
Original quote: βI find it often wrong and distracting and not any faster than I am.β
Management was sold by a financially invested speaker β Anonymous Authority (45/100)
At 6:00
He's calling out a 'speaker' who is just a mouthpiece for big money π°
Why this score: He attributes the company's shift to a generic 'speaker' whose only credential is being 'financially invested,' which is classic industry manipulation.
Original quote: βthere was a speaker who was, you know, heavily invested, literally financially invested in the future of AI and kind of sold all of the management-level people on AI is the futureβ
Sources: [Solved] Sean a retired CEO Invests capital In a startup company that - Strategic Management and Leadership (PADM-GP.2110) - Studocu, MGT 449 - Exam 1 Flashcards | Quizlet, BigSpeak Motivational Speakers Bureau - Keynote Speakers & Business Speakers
AI is mandatory for survival or you'll be obsolete via Loaded Language β Loaded Language (45/100)
At 6:30
Using 'obsolete' and 'left in the dust' to create a fear-based ultimatum π
Why this score: He's framing the entire industry as a binary choice between AI adoption and extinction. It's high-stakes drama for what is essentially just a tool update.
Original quote: βneeds to use AI or you will be left in the dust. You will be um obsolete essentially if you don't start making use of these tools.β
Personal attempt to use AI in Spring 2025 via Personal Story β Personal Story (60/100)
At 6:46
Just a guy sharing his timeline of trying to make it work π€·ββοΈ
Why this score: This isn't a claim about the world, just his own schedule. He's setting up his personal narrative arc.
Original quote: βI've had these bad experiences, but I'll try it. And this is would be in the spring of 2025.β
AI capability was surprisingly high in 2025 via Confidence Mismatch β Confidence Mismatch (45/100)
At 7:18
Vague 'surprising capability' without actual metrics or benchmarks π
Why this score: He's using subjective feeling ('it was neat') as a proxy for technical performance. 'Surprising' is doing all the heavy lifting here.
Original quote: βit was neat and and it was surprising how capable it was uh even at that time.β
AI speed outpaces human capacity via Emotional Button β Emotional Button (45/100)
At 8:08
Selling the 'high' of productivity without showing a single completed task π
Why this score: He's describing the dopamine hit of high-speed output. It sounds like a productivity miracle, but it's just him feeling the rush of automation.
Original quote: βI'm getting so much more done. Uh it's it's faster than I'm able to work, right?β
AI sounds too human, which is a major red flag π© β Emotional Button (45/100)
At 10:42
Using 'concerning' and 'worries me' to turn a technical observation into a moral crisis π
Why this score: He's pivoting from the technicality of conversational AI to the emotional weight of how non-experts will be manipulated by it. It's a vibe, not a data point.
Original quote: βI really think it's very concerning to me. Like, I I that worries me at like how people who aren't programmers use it.β
Personal fatigue after a year of 'agentic' coding π β Personal Story (60/100)
At 11:11
He's airing his own 'AI fatigue' after a year of trial runs. It's an experience, not a statistic π
Why this score: This is pure personal testimony about his own workflow. He's not claiming the industry is failing, just that he's personally exhausted by it.
Original quote: βI'm starting to have some doubts or I'm starting to like feel the um the impact of basically like being 6 to 7 to 8 or maybe almost this point 12 months into using agentic coding.β
AI coding tools are a massive, profitable business π° β No Frame (75/100)
At 12:13
Stating the obvious: companies are dumping money into these tools. No tricks here, just facts πΈ
Why this score: He's acknowledging the economic reality that drives AI adoption. It's a straightforward observation about market demand and corporate spending.
Original quote: βteams are willing to pay thousands of dollars a month for these tools. And that's a legitimate business need.β
Claims AI delegation has massive downsides via personal experience π β Personal Story (60/100)
At 12:31
He's pivoting from a tool to a 'downside' based on his own existential dread π
Why this score: He frames the shift to AI agents as a source of 'huge downsides' without defining them yet, relying on his own emotional response to justify the pivot.
Original quote: βWell, as I keep handing more and more off, you know, relegating implementation to the coding agent, there's huge downsides to this that I experienced.β
Claims developers will just become AI trainers via markdown files π© β Loaded Language (45/100)
At 13:07
Selling a future of 'training AI' as if it's the only path left π
Why this score: He uses a hypothetical 'role' to describe the future of the industry, making it sound like a hollowed-out version of actual engineering.
Original quote: βOh, well, the the role of the software development in the era of AI is to train the AI by creating different markdown files that basically prevent it from making the same mistakesβ
Claims AI-driven code review is an exhausting, high-volume chore π β Emotional Button (45/100)
At 13:38
Using the exhaustion of manual review to justify abandoning the tech π©
Why this score: He leans into the 'exhaustion' of reviewing massive amounts of AI code to build an emotional case for why the current workflow is broken.
Original quote: βAnd And like that's not what I like about coding. Basically, none of that. Code review is so difficult. Uh it takes a lot of time to do it really well.β
Claims AI-generated code leads to a lack of understanding π β False Dilemma (50/100)
At 14:24
Claims if you didn't write it, you don't own the knowledge. A bit of a reach π©
Why this score: He creates a binary where 'not writing it' equals 'not understanding it,' ignoring the skill of deep analysis and architectural oversight.
Original quote: βAnd you don't understand it cuz you didn't write it.β
AI agents cause developer apathy β personal anecdote β Personal Story (60/100)
At 14:56
He's making it about his own soul instead of the industry. Purely a vibe check π
Why this score: He explicitly frames this as his own feeling ('at least that's how I felt'), moving from a general observation to a personal emotional state.
Original quote: βAnd using these AI agents for coding creates this apathy towards the work and the software. At least that's how I felt.β
3x productivity doesn't boost revenue β anecdotal proof β Personal Story (50/100)
At 15:48
Claims 'proof' based on a single job. That's not data, that's just his Tuesday π
Why this score: He uses the word 'proof' to describe a single observation from his current workplace, which is an anecdotal leap rather than a statistical fact.
Original quote: βI have proof from my current job that being, let's say, three times more productive does not lead to any increase in revenue, right?β
Human-made software has an 'ineffable quality' β emotional argument β Loaded Language (45/100)
At 16:13
Using 'ineffable quality' and 'care' to sell a feeling. It's poetry, not engineering π
Why this score: He uses high-concept, emotional language like 'ineffable' and 'care' to make a qualitative argument that is impossible to measure objectively.
Original quote: βit's human-made by humans for humans with care. And it shows in the software.β
Engineers are losing core skills due to AI dependency π© β Loaded Language (45/100)
At 16:50
Using 'de-skilled' to turn a workflow shift into a tragedy π
Why this score: Heβs using heavy-handed terms like 'de-skilled' and 'hard programming work' to frame AI as a crutch that weakens the mind rather than just a tool.
Original quote: βAnd then you have an entire team of engineers who have basically de-skilled themselves. So, that if Claude or Cursor or whatever is down and there's an outage, you know, they they they haven't done hard programming work for 6 months or a year or three two years or however long.β
AI prompts use a 'scary' amount of water π β Anonymous Authority (45/100)
At 17:47
Claims 'scary' water usage without a single data point π
Why this score: He pivots from coding to environmentalism using vague, high-emotion descriptors like 'scary' and 'worrying' without citing a single study or metric.
Original quote: βOne, I start to learn more about the actual costs of AI. How much every prompt, how much water every prompt uses. An incredibly scary amount of water that that is being used.β
AI bubble is driving up hardware costs and causing 'immense harm' β Loaded Language (45/100)
At 18:30
Using heavy-hitting words like 'immense harm' and 'sucking everything up' to describe market fluctuations π
Why this score: He's making a massive, sweeping economic claim about hardware scarcity and 'harm' without citing specific market data or price indices. It's an emotionalized take on supply chain issues.
Original quote: βconsoles becoming more expensive, it being hard to find RAM for computers, like this AI bubble is just sucking everything up. Causing immense harm.β
AI-induced crisis leads to questioning entire career paths β Emotional Button (45/100)
At 19:59
Leaning into the 'existential crisis' to drive home how much this tech affects his soul π
Why this score: He's pivoting from technical growth to a heavy emotional crisis about career identity. It's an effective way to end, but it moves from 'coding stats' to 'soul-searching' very quickly.
Original quote: βSo combine all of that and feeling like I'm having this crisis of like, should I become an electrician or should like a nurse...β
Claims the industry is forcing AI usage through sheer pressure β Loaded Language (45/100)
At 20:35
Uses heavy emotional weight like 'guilt' and 'being told I have to' to frame a technical choice as a moral crisis π
Why this score: The speaker frames the adoption of AI not as a tool choice, but as an inevitable weight they are being crushed by. It's high-drama framing for a workflow decision.
Original quote: βAnd I'm being told I have to use it. And I just thought, well, what do I really have to have to use it? What if I just stop using it?β
Predicts total career collapse as a consequence of rejecting AI β Emotional Button (45/100)
At 21:24
Leans into the 'all or nothing' stakes to make a personal decision feel like a grand sacrifice π
Why this score: He's playing the martyr card. While losing a job is possible, claiming his entire career is 'over' if he skips AI is an emotional gamble to justify a lifestyle change.
Original quote: βif I get fired, if I can't find a job, if my career as I know it is over, so be it.β
Makes a sweeping claim about needing to avoid AI entirely β False Dilemma (45/100)
At 21:43
Presents the choice as 'use AI for everything' or 'use nothing at all'βignoring the middle ground π©
Why this score: Heβs creating a binary: you either succumb to the machine or live like a hermit. It ignores the reality of hybrid workflows that most developers actually use.
Original quote: βIt's not using AI for coding. It's not using AI for anything.β
See the full analysis with sources and timestamps β