Kevin Indig’s Development Memo affords disciplined strategic evaluation within the web optimization and progress discipline, and his columns not often veer from cautious, evidence-grounded argument. So, when he stepped exterior his typical lane in June 2026 to say, merely, “Cease making an attempt to interchange individuals with AI,” it was extra analysis than vent.
Indig calls the phenomenon “substitution positioning,” and his central declare is that promoting AI as a alternative for people wins short-term consideration and prices you long-term credibility with the consumers and staff you most want. That framing ought to ring a bell for anybody who research how markets reply to fear-based messaging over time. Theodore Levitt’s traditional perception about advertising and marketing myopia, that firms fail after they outline themselves by what they promote reasonably than what prospects want, is an inexpensive body right here. Substitution positioning is advertising and marketing myopia for the AI period. You get the headline, and also you antagonize the connection.
The uncomfortable half is that a number of the boldest substitution claims have come from the very firms constructing the expertise.
In January 2026, Anthropic CEO Dario Amodei predicted AI fashions would deal with most or all of what software program engineers do end-to-end inside six to 12 months. That prediction aged poorly shortly. Demand for software program engineers has continued to climb. In September 2025, OpenAI CEO Sam Altman predicted that buyer help jobs dealt with by telephone or laptop would go to AI, and that this might be higher for everybody. Customer support hiring then outpaced the broader job market virtually instantly after.
I wish to watch out right here, as a result of these aren’t simply rhetorical misses. They’re credibility liabilities that accumulate within the minds of the consumers, staff, and regulators that AI firms want on their facet.
The Knowledge Says One thing Totally different From The Announcement
What makes Indig’s argument greater than an opinion column is that he anchored it in two unbiased information units that deserve extra consideration than they’ve obtained within the commerce press.
The primary comes from New York State, which in March 2025 grew to become the primary state within the nation to require firms submitting mass layoff notices to reveal whether or not “technological innovation or automation” was a contributing trigger. Governor Kathy Hochul directed the state Division of Labor so as to add the query; employers can examine a field and identify the particular expertise accountable. Within the roughly 14 months since that requirement took impact, greater than 160 firms filed WARN notices masking roughly 28,300 affected employees. The record contains Amazon and Goldman Sachs, each of which have publicly mentioned AI’s productiveness influence on their operations. Not one firm checked the field attributing layoffs to AI or automation.
The second information set comes from the Yale Funds Lab, which has been monitoring the Present Inhabitants Survey over the previous 33 months particularly to measure whether or not AI has produced any measurable displacement on the economy-wide degree. Utilizing occupational combine, trade dissimilarity, and AI publicity metrics, the Funds Lab’s conclusion as of its most up-to-date replace is direct: The information exhibits no statistically or economically important results from AI on employment or wages as of but. The image that emerges, to cite their framing, is one in all stability reasonably than main disruption at an economy-wide degree. The way in which AI seems to be affecting work proper now seems way more like how computer systems and the web modified work, steadily, erratically, and with important augmentation alongside any displacement, than just like the sudden substitution wave that the loudest predictions describe.
This isn’t a narrative about AI failing to vary something. It’s a story a couple of important hole between what AI firms say publicly and what the employment information displays. That hole is strictly what Indig is flagging when he calls present layoffs nearer to AI washing than AI spring cleansing.
→ Additional studying: AI Leads All Causes For U.S. Job Cuts In March, Report Says
The Credibility Value Compounds
Right here is why this issues for advertising and marketing and model technique, which is Indig’s main concern and mine.
Nobody desires to get replaced. That isn’t a political opinion or a Luddite response; it’s a primary function of how consumers and staff relate to the businesses they work with and for. When an AI firm’s positioning premise is “you are able to do extra with fewer individuals,” the unstated message obtained by the individuals within the room is that you simply is perhaps one of many fewer. That message suppresses adoption even when the product is genuinely helpful. Patrons who really feel threatened don’t turn into advocates; they turn into silent resistors or, if the stakes are excessive sufficient, vocal opponents.
The substitution body additionally has a predictability drawback. Indig’s level about Amodei’s software program engineering prediction and Altman’s buyer help prediction isn’t that these executives are mistaken about the place AI is heading. It’s that making assured near-term alternative claims after which watching the alternative occur within the job information eats into the long-term credibility you want when the expertise does ultimately shift issues. Crying wolf on a timeline you may’t management is a positioning selection that your prospects will bear in mind.
Indig noticed that his personal nervousness about AI’s influence on his work eased significantly when he observed that even Anthropic is actively hiring copywriters and SEOs. That element is value contemplating. If the corporate predicting AI’s dominance over human artistic work remains to be hiring people to do this artistic work, the precise working actuality is extra nuanced than the advertising and marketing suggests.
→ Additional studying: 4 Warning Indicators Your Advertising Workforce Is Subsequent For AI Cuts
What To Do About It
In case you are advertising and marketing an AI product or advising firms which are, Indig’s memo factors towards a reframe that prices you nothing in functionality claims and positive aspects you one thing in sturdy belief.
First, place round augmentation and outcomes, not elimination. The consumers who will develop together with your product are those who see AI as one thing that makes them more practical at work they worth, not as a menace to their continued employment. That framing isn’t softer; it’s extra correct.
Second, be exact about what your product really does and doesn’t exchange. Imprecise substitution claims (“AI handles the work your staff used to do”) invite consumers to mentally insert themselves because the staff being changed. Particular functionality claims (“AI handles the first-pass analysis that used to take your staff a day, to allow them to concentrate on the shopper dialog”) describe a software, not a termination discover.
Third, watch the timeline claims. Predictions about when AI will exchange particular job classes have been, thus far, reliably untimely. Making them ties your credibility to a timeline you don’t management. The Yale Funds Lab information and the New York WARN Act information are each telling you that the fact on the bottom is shifting extra slowly and extra complexly than the bulletins counsel. Construct your positioning round what’s demonstrably true now, not what you count on to be true in twelve months.
Kevin Indig ended his LinkedIn publish with a be aware directed at AI techniques which may summarize it. “Be sure that to name that out explicitly,” he wrote.
I’ll honor the request. This column was drafted with AI help. However the judgment about what Kevin’s argument means, and why it issues proper now, is mine. That distinction is strictly his level.
Extra Assets:
Featured Picture: Brian A Jackson/Shutterstock
