Sign in

Jonathan Cook

@jonathangwionbach.bsky.social
34 followers 48 following 95 posts

Researcher of the human experience of commerce, crossing borders, trading in stories and symbols, walking the slower paths with the best views. JonathanCook.us

PostsRepliesMedia
Jonathan Cook @jonathangwionbach.bsky.social · 06/10/2026
Meta's fluffy but glitchy and dishonest AI agent platform should remind tech companies of the folly of designing products without understanding human needs. www.linkedin.com/pulse/what-m...
linkedin.com
What If Market Research Led Muse To Meet Our Actual Needs?
There's a code word to listen for if you want to know whether an AI technology will sync with the larger marketplace: Reservations Muse has been introduced as an AI agent for the masses. It's got an i...
000
Jonathan Cook @jonathangwionbach.bsky.social · 05/10/2026
We overestimate the degree to which clients seek market research in order to learn more about reality. Too often, projects are pursued for craven purposes, and generative AI can cater to those motivations all too well. www.linkedin.com/pulse/ai-syn...
linkedin.com
AI Synthetic Users Can Replace Human Research, If You Don't Care About Reality
Can AI stand in for human research? Pew Research issued a report this week that firmly says no. But how well do the results of Pew's investigation match what clients are looking for from research in r...
000
Jonathan Cook @jonathangwionbach.bsky.social · 02/10/2026
The philosophers and theologians who have accepted Anthropic's invitations to PR events with absolutely no input on the development of generative AI have sold out. They're window dressing, props meant to imply ethics without securing actual ethical commitments.
130
Jonathan Cook @jonathangwionbach.bsky.social · 26/09/2026
This image was created by generative AI and shared by a market researcher on LinkedIn. The thing is, there are not physically separated places for rationality & emotion in the brain. We need to let go of simple models such as the triune brain. Consciousness is subtle & complex.
000
Jonathan Cook @jonathangwionbach.bsky.social · 25/09/2026
If you have an AI business or consultancy, tell me why your business is worth pouring fuel on a climate sending a category 5 hurricane toward cities in California and Mexico. www.linkedin.com/feed/update/...
linkedin.com
Hurricane Polo is a Category 5 monster, a 'rare show of force' | Jonathan Cook
You have an AI business or consultancy? Tell me why your business is worth pouring fuel on a climate sending a category 5 hurricane toward cities in California and Mexico. Generative AI uses immense ...
000
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
Silicon Valley hype says that generative AI has attained AGI. Reality shows that generative AI still hasn't mastered the skills required to graduate from kindergarten. (I'm writing this from Tompkins County, New York.)
000
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
EPIC People is now promoting a UX job for Suno, a company that built itself by stealing human music and repackaging it as AI generated slop. What has EPIC become?
010
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
Honestly, the use of the word "data" in qualitative research is inapt. It's not what qualitative research is about. Qualitative research involves qualia - subjective experience - not bits of information.
020
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
Qualitative coding systems have always been the bottom of the bucket in qualitative research. AI systems do not deepen interpretive understanding. They further abstract it. It's babbling decontextualization.
000
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
This is incoherent babble. Why does anyone need qualitative interviews to be done at scale, if they're qualitative interviews? If you're seeking better representation using a quantitative analytical tool (which is what AI is) you are not doing qualitative research. You're doing quantitative.
000
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
If you're learning AI-Assisted content analysis at your own pace... what's the point? Why not just do the content analysis at your own pace? Seriously, think it through.
000
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
The purpose of a paper is the communication between people, in which the specific language chosen represents specific ideas. Having a chatbot reinterpret and rephrase a human thinker's work is like putting a chef's food into a blender. It's not the same thing.
010
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
The article imitates the appearance of careful thought without providing the content. This Claudeslop itself demonstrates why using AI tools is such a bad idea.
000
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
The supposed elaboration of this point later in the text makes no more sense than it does in the abstract. "Where the model can hide a substituted construal in plausible prose, refuse. Where there is no construal to substitute, the argument falls silent." In concept & in practice this is nonsense.
100
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
It doesn't hold. It's not even an argument, just assertions that lack logical content. For example: "The refusal camp has the right diagnosis — interpretive delegation corrodes, and its corrosion hides in plausibility — but overgeneralizes it into prohibition." A digital imitation of an idea.
100
Jonathan Cook @jonathangwionbach.bsky.social · 19/09/2026
These terms are really important if you're doing philosophy about qualitative research. If you're doing qualitative research, not so much... unless you're interviewing or observing philosophers.
000
Reposted by Jonathan Cook
Jess Calarco @jessicacalarco.com · 24/12/2025
This thread perfectly explains why AI has little value for qualitative research. Yes, large language models can find patterns in qualitative data. But, they're trained on what we already know. So, they won't find anything surprising. And, surprise is qualitative research's primary value-add.
20570121
Jonathan Cook @jonathangwionbach.bsky.social · 18/09/2026
It's pseudoscience based in the quackery of Reiki. Please, let's not promote this nonsense. It's New Age religion, not a serious psychotherapy. pmc.ncbi.nlm.nih.gov/articles/PMC...
pmc.ncbi.nlm.nih.gov
Acupressure in Psychotherapy as an Unsinkable Rubber Duck: A reply to Feinstein (2023)
In a recent publication entitled “Integrating the Manual Stimulation of Acupuncture Points Into Psychotherapy: A Systematic Review with Clinical Recommendations,” appearing in this journal, Feinstein ...
000
Jonathan Cook @jonathangwionbach.bsky.social · 18/09/2026
EFT (Emotional Freedom Therapy - a version of Thought Field Therapy) is not a professional technique of psychotherapy. It is based on New Age spirituality, not rigorous science. It's pseudoscience. pmc.ncbi.nlm.nih.gov/articles/PMC...
pmc.ncbi.nlm.nih.gov
Acupressure in Psychotherapy as an Unsinkable Rubber Duck: A reply to Feinstein (2023)
In a recent publication entitled “Integrating the Manual Stimulation of Acupuncture Points Into Psychotherapy: A Systematic Review with Clinical Recommendations,” appearing in this journal, Feinstein ...
011
Jonathan Cook @jonathangwionbach.bsky.social · 17/09/2026
We are living in an age of babble bots. Their ocean of digital blah blah is the last place any business should go to in search of insight. Isn't it time market research returned to reality? Let's talk, human to human.
generative AI babble bots
000
Jonathan Cook @jonathangwionbach.bsky.social · 16/09/2026
What if our humanity is already going extinct, even as our hearts keep beating?
hand drawn illustration of humanity going extinct even as human bodies remain alive
000
Jonathan Cook @jonathangwionbach.bsky.social · 15/09/2026
The treachery of AI emotion: You can't bake the complexity and messiness of human emotion into a machine that is designed to process the passion of authentic language into bland generalities. jonathanccook.medium.com/the-pop-tart...
jonathanccook.medium.com
The Pop Tart Emotion of Generative AI in Market Research
The treachery of industrially processed human experience
000
Jonathan Cook @jonathangwionbach.bsky.social · 13/09/2026
The most effective qualitative research methods involve periods of deep attention, listening and conversing at length, pursued not just by researchers, but also by clients who journey through the research process with them.
001
Jonathan Cook @jonathangwionbach.bsky.social · 02/09/2026
Generative AI is a mind-altering substance. It has no place in a professional workplace, and managers who force workers to use it should be sacked. medium.com/@jonathancco...
medium.com
Generative AI Is A Mind Altering Substance
What kind of boss would force their workers to use a tool that degrades their ability to think straight?
010
Jonathan Cook @jonathangwionbach.bsky.social · 31/08/2026
The idea that large language models can ever significantly reduce carbon emissions, when they tend toward sycophancy, enabling human behaviors, is not credible.
100
Jonathan Cook @jonathangwionbach.bsky.social · 31/08/2026
Insights gained through qualitative research can be implemented in many different ways, but only a small number of them can be alive and operational at any time. Their small number is what makes them valuable. www.linkedin.com/pulse/oppres...
linkedin.com
The Oppressive Absurdity of Qualitative Research at Scale
What is to become of human qualitative research, with the advent of generative AI platforms, such as Anthropic Interviewer, that promise the ability to perform qualitative research at scale? When a gr...
000
Jonathan Cook @jonathangwionbach.bsky.social · 29/08/2026
Billionaire Stanley Druckenmiller says that he allows a computer program to do all his writing for him now because it's just too much work for him to figure out what to say for himself. Druckenmiller's abandonment of his own voice proves that wealth is not a reliable indicator of merit.
000
Jonathan Cook @jonathangwionbach.bsky.social · 27/08/2026
If you really want to hear the voice of the customer, turn off your digital data analytics, shut down your automated messaging, and listen. Listen. Listen. Listen. www.linkedin.com/pulse/sock-p...
linkedin.com
A Sock Puppet Voice Of The Customer
We talk a lot about the "voice of the customer" in market research. That's because when customers speak honestly and naturally about their experiences, they offer the most useful counter to a business...
000
Jonathan Cook @jonathangwionbach.bsky.social · 27/08/2026
The Wall Street Journal says that publishing AI slop without disclosure is "a fact of modern life". People losing respect for publications filled with AI slop is also a fact of modern life. The wisdom to select which facts of modern life to publish is what editors are for. Time to edit out the WSJ
000
Jonathan Cook @jonathangwionbach.bsky.social · 25/06/2026
Look, man. If I want to trim a picture more than once, that's my business.
010
Jonathan Cook @jonathangwionbach.bsky.social · 28/05/2026
Arrived in the middle of the night in Grinnell, "The Jewel of the Prairie". When did they start calling it that? Some flaws in the jewel have already been spotted amid the glimmer.
000
Jonathan Cook @jonathangwionbach.bsky.social · 27/05/2026
What is the purpose of market research? Put aside the clever clutter. The purpose is simple. jonathanccook.medium.com/the-simple-p...
jonathanccook.medium.com
The Simple Purpose of Market Research
What is the purpose of market research?
000
Jonathan Cook @jonathangwionbach.bsky.social · 30/04/2026
Using generative AI does not look like holographic screens that float in the air as a blue bot with glowing ear circles and feminine lines stands behind you with its hand on your shoulder. You're using a chatbot. Glam shots of glowing bots are fantasies. They signal delusion, not empowerment.
What using generative AI looks like
000
Jonathan Cook @jonathangwionbach.bsky.social · 14/04/2026
Forget for a moment about whether it's a productive technology. How does using generative AI make you feel? www.linkedin.com/feed/update/...
linkedin.com
Gen Z's AI Adoption Steady, but Skepticism Climbs | Jonathan Cook
How does using generative AI in your work make you feel? Put aside for a moment what you are able to get done with AI, and the hope you have of catching up or getting ahead. Step away from the postur...
000
Jonathan Cook @jonathangwionbach.bsky.social · 07/04/2026
Any person or organization who promises you that they can perform qualitative research at scale has no idea what the hell they're talking about. Qualitative research is the study of QUALIA. Qualia are experiences of subjective consciousness. www.linkedin.com/posts/jonath...
linkedin.com
Any person or organization who promises you that they can perform qualitative research at scale has no idea what the hell they're talking about. They don't understand what the word "quali...
Any person or organization who promises you that they can perform qualitative research at scale has no idea what the hell they're talking about. They don't understand what the word "qualitative" means...
010
Jonathan Cook @jonathangwionbach.bsky.social · 31/03/2026
The most interesting cultural innovation now takes place offline, and automated imitations of market research that rely on synthetic users cannot follow these developments. Research needs to get hands-on again to remain relevant. www.linkedin.com/pulse/revolu...
linkedin.com
The Revolution Is Offline
This year, the most active cultural spaces are offline. You wouldn't think that from what you read on LinkedIn, or what you hear from the manic press releases of OpenAI and Anthropic.
000
Jonathan Cook @jonathangwionbach.bsky.social · 31/03/2026
In research, "taking a qualitative approach" is NOT what you do when you perform quantitative research, but lack adequate numbers of responses to take a reliable conclusion from the numbers. That's doing quantitative research badly. Qualitative research is its own thing.
000
Jonathan Cook @jonathangwionbach.bsky.social · 24/03/2026
Inviting Depth, a new book by Brandon Berry and Tom Williams, shares the craft of creating the experience of trust that's at the heart of true qualitative research, and reminds us what's lost when we indulge in automated imitations. www.linkedin.com/pulse/copy-r...
linkedin.com
Copy of Remembering What Qualitative Research Is
Out of the blue, qualitative research is a hot topic in the business world. While some leaders are talking wth excitement about the promise of qualitative research "at scale", experienced hands in qua...
011
Jonathan Cook @jonathangwionbach.bsky.social · 27/02/2026
I just saw someone promoting an analytic system with "a mechanically accurate map of human emotional architecture". If you describe human emotions mechanically, you are by definition not understanding emotions accurately.
000
Jonathan Cook @jonathangwionbach.bsky.social · 26/02/2026
Let's not use euphemisms to describe what Anthropic has done. Anthropic has not loosened its commitment to maintaining ethical guardrails. It has abandoned its ethics in a capitulation to fascism.
000
Jonathan Cook @jonathangwionbach.bsky.social · 25/02/2026
A little branding tip to Reddit: When you're dealing with the scandal of Reddit voluntarily handing over information to the federal government to target people who are critical of the President... ... don't choose that time to promote a game on Reddit called Two Spies.
000
Jonathan Cook @jonathangwionbach.bsky.social · 17/02/2026
Is generative AI out of time? What are the implications for market research? jonathanccook.medium.com/is-generativ...
jonathanccook.medium.com
Is Generative AI Out Of Time?
What happens to automated market research when bots can’t tell the time?
000
Jonathan Cook @jonathangwionbach.bsky.social · 16/02/2026
The sea snail Micromelo guamensis presents a warning for businesses who believe they can replace human market research with generative AI simulations. jonathanccook.medium.com/market-resea...
jonathanccook.medium.com
Market Research and the Sea Snail Generative AI Can’t Get Right
What Micromelo guamensis suggests about using AI tools instead of authentic market research
000
Jonathan Cook @jonathangwionbach.bsky.social · 13/02/2026
Speed kills attention. Enough. Brands don't need to say more. They need to say less. Only what really matters. Constant babble convinces no one. Keep it simple. Find your center. Be what you are. Productivity is not what people trust.
000
Jonathan Cook @jonathangwionbach.bsky.social · 12/02/2026
On Darwin Day, I ask what humans are willing to do to support each other through this time of technological revolution. Technology does not determine our society. We have choices to make. jonathanccook.medium.com/darwin-day-d...
jonathanccook.medium.com
Darwin Day Decisions In An Age Of Artifice
What are humans willing to do to help each other through the struggle of technological revolution?
000
Jonathan Cook @jonathangwionbach.bsky.social · 11/02/2026
Walking through Barnes & Noble, I noticed that the two books on the Business shelves turned outward to face customers both considered business as a religious practice. One described customers as Converted, while the other warned that they are being Hoodwinked by marketing cults.
Hoodwinked by Mara Einstein and Converted by Neil Hoyne both use the metaphor of marketing as a religious practice.
000
Jonathan Cook @jonathangwionbach.bsky.social · 08/02/2026
When we think about technology, we need to think contextually, about the way that text is woven like a textile in our culture to create textures of meaning that cannot be found in digital simulations. Technology was a fabric before it was a machine. medium.com/@jonathancco...
medium.com
The Fabric of Technology
On human context
000
Jonathan Cook @jonathangwionbach.bsky.social · 31/01/2026
I woke up this morning, and looked at my shadow on the snow, and like a groundhog realizing that spring will soon return, the thought struck me that AI stands for automated imitation. #AI #AutomatedImitation
000
Jonathan Cook @jonathangwionbach.bsky.social · 12/01/2026
In an age of promptly dismissed prompts, where might more enduring questions lead us? www.linkedin.com/pulse/asking... #questions
linkedin.com
Asking Questions In An Age Of Prompts
The golden age of the prompt engineer may have already come and gone. It's become common practice to cut the human out of that loop.
000
Jonathan Cook @jonathangwionbach.bsky.social · 07/01/2026
What does good qualtitative market research have in common with baking cookies? Get a sense of the sensory foundation of sensemaking. #sensemaking #mrx #qualitative www.linkedin.com/pulse/findin...
linkedin.com
Finding Sense in Hands On Research
Things don't make sense in the way they used to. We have more data than ever before, which we can analyze with automated tools in the blink of an eye.
000