Contact Explore Products

Tag: generative AI

  • AI Is Making Us Dumber — and the Data Finally Proves the Productivity Trap

    AI Is Making Us Dumber — and the Data Finally Proves the Productivity Trap

    AI helped people perform better while they were using it.

    Then the AI disappeared—and their performance got worse.

    That is the uncomfortable result of a peer-reviewed study involving nearly 1,000 high-school students. With access to a standard GPT-4 assistant, students performed 48% better during practice than students working without AI. But when they took an exam without assistance, they performed 17% worse than the control group.

    The tool improved the visible output while weakening the learning underneath it.

    That should concern more than teachers and parents.

    Every day, marketers ask AI to create positioning. Ecommerce teams use it to analyse reviews, write product pages and design experiments. Managers ask it to summarise documents they have not read. Developers accept code they could not explain five minutes later.

    The work gets finished. The dashboard looks productive. But are we becoming more capable—or merely more dependent?

    Welcome to the AI productivity trap.

    First, what did the study actually prove?

    The paper, Generative AI Without Guardrails Can Harm Learning, was published in the journal Proceedings of the National Academy of Sciences.

    Researchers conducted a randomized controlled trial at a large high school in Turkey during the 2023–2024 academic year. Nearly 1,000 students across grades 9, 10 and 11 participated in four 90-minute mathematics sessions. Classes were assigned to one of three groups:

    1. Control: Students used their normal books and notes without generative AI.
    2. GPT Base: Students had a standard GPT-4 chat interface similar to ChatGPT.
    3. GPT Tutor: Students used GPT-4 with teacher-designed safeguards. It was instructed to provide hints, ask students to show their work and avoid giving away complete answers.

    Each session contained assisted practice followed by an exam completed without AI or other resources.

    The results were striking:

    • GPT Base improved assisted practice performance by 48% compared with the control group.
    • GPT Tutor improved assisted practice performance by 127%.
    • On the unassisted exam, the GPT Base group performed 17% worse than the control group.
    • The GPT Tutor group performed about the same as the control group on the exam. Its learning penalty was essentially removed, but it produced no statistically detectable exam improvement.

    The researchers also examined how students interacted with the systems. GPT Base users frequently asked for answers and copied solutions. GPT Tutor users were more likely to attempt answers, ask for help and work through the problem.

    Perhaps the most worrying finding was that students did not recognise the damage. Those using the standard assistant did not believe they had learned less or would perform worse.

    Their output gave them a feeling of mastery that their independent performance could not support.

    Important fact check: this does not prove AI lowers intelligence

    “AI is making us dumber” is a deliberately provocative headline. The study did not measure IQ, permanent cognitive decline or the long-term impact of AI on adults at work.

    It studied short-term mathematics learning in one Turkish high school. The exam followed the practice session, and the authors explicitly noted that long-term learning remains a subject for future research.

    We should therefore avoid turning one strong experiment into a universal law.

    What the study does demonstrate is narrower—and highly relevant:

    AI can increase performance while the tool is present without building the user’s underlying ability. When the tool provides answers too easily, independent performance can become worse.

    That is not proof that AI makes everyone less intelligent. It is evidence that unguarded AI use can replace the mental effort required to learn.

    For businesses, that distinction is more useful than the headline.

    The difference between performance and capability

    We often treat these words as if they mean the same thing.

    They do not.

    Performance is the quality of what you produce today.

    Capability is what you can understand, judge and produce tomorrow—even when the tool is unavailable, wrong or facing a situation it has never seen.

    AI can raise performance instantly. It can write a polished email, summarise a report, produce ten campaign ideas or generate a complete product description in seconds.

    But polished output can hide weak understanding.

    If you cannot explain why a recommendation is correct, recognise when it is wrong or reproduce the reasoning in a new context, the capability may belong to the tool rather than to you.

    That is the trap: borrowed competence feels like personal competence while the AI is present.

    How the AI productivity trap appears in marketing

    Imagine a marketer asks AI:

    “Create a campaign strategy for our new skincare product.”

    The model produces personas, hooks, channel recommendations and a 30-day content calendar. The result looks comprehensive. The marketer cleans up the language, puts it into slides and presents it.

    But ask three follow-up questions:

    • Why is this the right audience?
    • Which customer evidence supports this positioning?
    • What would make you abandon this strategy after launch?

    If the marketer cannot answer without reopening the chatbot, AI did not multiply their strategy. It substituted for it.

    Over time, this changes the job. The marketer becomes skilled at requesting and formatting answers but less practised at customer research, positioning and making trade-offs.

    The deliverables continue. The thinking weakens quietly.

    How it appears in ecommerce

    The same risk exists across ecommerce operations.

    Product descriptions

    AI can create hundreds of descriptions quickly. But if nobody understands the customer’s objections, product differentiators or regulatory boundaries, the catalogue becomes fluent and generic.

    Customer-review analysis

    An AI summary may identify recurring complaints. But the operator who never reads the original reviews may miss sarcasm, emerging edge cases or the emotional language customers actually use.

    Conversion optimisation

    AI can recommend tests, but it does not automatically understand your traffic quality, technical constraints, commercial margins or brand history. A team that accepts experiments without forming its own hypothesis is generating activity, not learning.

    Reporting

    AI can explain why revenue changed. If the analyst cannot trace that explanation back to reliable data, it becomes a confident story attached to a chart—not analysis.

    In each case, the immediate task becomes easier. The organisation’s ability to question, interpret and decide can still become weaker.

    The AI Crutch Test

    You do not need to stop using AI. You need a quick way to recognise when assistance has become dependence.

    Ask yourself these three questions.

    1. Could I explain the result without reopening the AI?

    You do not need to reproduce every sentence from memory. You should be able to explain:

    • What conclusion was reached
    • What evidence supports it
    • Which assumptions it depends on
    • Where it might fail

    If you can only defend the output by saying “the AI suggested it,” you have an answer but not understanding.

    2. Did I think before I prompted?

    Write down your initial hypothesis, outline or decision criteria before asking the model.

    For a campaign, define the customer, problem and desired action. For an ecommerce test, state what you expect to happen and why. For research, list what evidence would change your mind.

    Then use AI to challenge, expand or improve that thinking.

    If the blank prompt box is always the beginning of your thought process, you are outsourcing the most valuable part.

    3. Can I detect a confident but wrong answer?

    The study’s ordinary GPT interface sometimes produced incorrect mathematics. Yet the researchers found that copying—not merely exposure to incorrect answers—was the main mechanism behind the learning penalty.

    The lesson is important: AI errors are most dangerous when the user has stopped engaging deeply enough to notice them.

    Before using an output, ask:

    • Which claims require verification?
    • Does this conflict with our data or experience?
    • What information might the model be missing?
    • Would I be comfortable putting my name behind this decision?

    If you lack enough domain knowledge to evaluate the answer, involve someone who has it.

    If you answer “no” to any part of the AI Crutch Test, change how you are using the tool—not necessarily the tool itself.

    Use AI as a coach, not an answer machine

    The most encouraging part of the study was the guarded GPT Tutor.

    It used the same underlying GPT-4 technology, but its behaviour was different. It had correct, teacher-supplied problem information. It was instructed to give hints rather than full solutions and to ask students to attempt the work first.

    That design eliminated the detectable exam penalty.

    For professionals, we can reproduce some of those safeguards through better workflows.

    Instead of asking:

    “Write the strategy.”

    Try:

    “Here is my proposed strategy and the customer evidence behind it. Challenge my assumptions, identify the weakest argument and ask me three questions before recommending changes.”

    Instead of:

    “Analyse this performance report.”

    Try:

    “I believe conversion fell because mobile traffic shifted toward a lower-intent source. Test this hypothesis against the data, show contradictory evidence and separate facts from inference.”

    Instead of:

    “Give me the answer.”

    Try:

    “Do not solve this immediately. Give me one hint, let me attempt it, then critique my reasoning.”

    The goal is to keep yourself inside the cognitive loop.

    A practical framework: Think, Ask, Verify, Rebuild

    For work that develops an important professional skill, use this four-step process.

    Think

    Form your first view without AI. Even five minutes of independent thinking forces you to retrieve knowledge, notice uncertainty and create a position worth testing.

    Ask

    Use AI for counterarguments, options, missing evidence and feedback—not only for finished deliverables.

    Verify

    Check material claims against original sources, company data and domain experts. Separate what the model knows from what it inferred.

    Rebuild

    Close the AI and summarise the conclusion in your own words. State the decision and why you made it. If you cannot, you are not finished.

    This takes longer than copying the first response. That small amount of productive friction is the point.

    Not every task needs to teach you something

    There is an important counterargument: we routinely use calculators, search engines, templates and automation without practising the underlying task every time.

    That is sensible.

    You do not need to preserve your ability to manually rewrite 500 product titles or format meeting notes. Offloading low-value work is one of AI’s biggest benefits.

    The real question is whether you are automating execution or surrendering judgement.

    Use AI aggressively for:

    • Reformatting and repetitive transformations
    • First-pass transcription and summarisation
    • Translation drafts followed by appropriate review
    • Generating variations after the core strategy is decided
    • Routine documentation

    Keep humans actively involved in:

    • Defining the problem
    • Evaluating evidence
    • Making strategic trade-offs
    • Understanding customers
    • Approving high-impact decisions
    • Recognising when the situation has changed

    The boundary will differ by role, but every team should define it intentionally.

    The biggest risk is the skill you stop practising

    AI dependence will not always feel like decline.

    It may feel like speed.

    You complete more tasks, respond faster and produce work that looks more polished. Meanwhile, the moments that once forced you to struggle, remember, compare and decide slowly disappear from your day.

    That struggle was not always inefficiency. Sometimes it was the mechanism through which expertise developed.

    The study’s students were not lazy or unintelligent. They reacted naturally to a system that offered an easier path to the answer. Professionals will do the same—especially when every workplace metric rewards output today rather than capability next year.

    That is why individual discipline is not enough. Leaders must design workflows and incentives that preserve thinking.

    Do not ask only, “How much faster did AI make the team?”

    Also ask:

    • Can the team explain and defend the output?
    • Are people improving at their core discipline?
    • Can they work when the model fails?
    • Is AI producing learning—or merely deliverables?

    AI should make your judgement more powerful, not less necessary.

    Final thought

    The future will not belong to people who refuse AI. It will not automatically belong to the people who use it most either.

    It will belong to those who know when to delegate to AI, when to challenge it and when to close the window and think for themselves.

    Use the tool to increase your range.

    Just make sure the intelligence in the workflow still includes yours.

    Frequently asked questions

    Does research prove that AI is making people dumber?

    No. The cited study did not measure intelligence or permanent cognitive decline. It found that unguarded GPT-4 assistance improved immediate practice performance but reduced short-term unassisted exam performance in a specific high-school mathematics setting.

    What was the 48% and 17% AI study?

    In a randomized trial involving nearly 1,000 Turkish high-school students, a standard GPT-4 assistant increased performance during assisted practice by 48% relative to the control group. The same group later performed 17% worse than the control group on an immediate exam without AI.

    Did the guarded AI tutor harm learning?

    The guarded GPT Tutor improved assisted practice performance by 127% and performed approximately the same as the control group on the unassisted exam. It removed the statistically detectable penalty but did not create a statistically significant exam improvement.

    How can professionals prevent AI skill atrophy?

    Think before prompting, use AI to challenge rather than replace your reasoning, verify important claims and explain the final decision without relying on the original AI response.

    Should marketers and ecommerce teams stop using AI?

    No. They should automate repetitive execution while retaining human ownership of customer understanding, strategy, evidence evaluation and final decisions.

    Sources

  • AI Companies Are Destroying Physical Books. Here’s Why Your Business Should Care.

    AI Companies Are Destroying Physical Books. Here’s Why Your Business Should Care.

    Imagine spending years writing a book.

    Then imagine an AI company buying a second-hand copy, slicing off its spine, scanning every page and sending the remains for recycling. The words survive—but now as data inside a private system built to generate commercial products.

    This is not a dystopian thought experiment. Court records show that Anthropic, the company behind Claude, bought and destructively scanned millions of print books while building an internal digital library and training its AI models.

    The easy reaction is outrage: A technology company destroyed books to build a machine that writes.

    But for creators, ecommerce teams and business owners, the more useful question is this:

    If AI companies can treat physical knowledge as a resource to acquire, process and discard, how should you expect them to treat your website, product descriptions, customer conversations and creative work?

    That is the part every business should be thinking about.

    What actually happened?

    According to documents disclosed in the US copyright case Bartz v. Anthropic, Anthropic created a large internal collection of books from two very different sources.

    First, it downloaded more than seven million books from pirate websites. Second, it legally bought millions of printed books and converted them into digital files.

    The physical process was destructive by design. Books had their bindings or spines removed so that loose pages could pass through high-speed scanners. Court filings described industrial cutting equipment, production scanners and recycling of the paper after digitisation.

    In a June 2025 ruling, US District Judge William Alsup treated those two routes differently:

    • Converting legally purchased print books into internal digital replacements was held to be fair use in this case.
    • Acquiring and retaining pirated copies for a general-purpose library was not excused as fair use.
    • Training on the works was also held to be transformative on the record before the court.

    That distinction matters. “The court said AI companies can steal books” is not an accurate summary. The ruling separated lawful purchase and format conversion from the acquisition of pirated material.

    Anthropic later agreed to a $1.5 billion settlement concerning pirated books, without admitting wrongdoing. The settlement did not erase the court’s earlier fair-use ruling on training and the destructive scanning of lawfully purchased copies.

    Were rare books really destroyed?

    This is where the viral version of the story often runs ahead of the evidence.

    It is confirmed that Anthropic destructively scanned millions of purchased books. It is also true that booksellers in several countries have reported strange bulk orders containing obscure, old and out-of-print titles. Some sellers suspect those orders are connected to AI training and that the books may be pulped after scanning.

    However, there is not yet public proof that AI companies are systematically targeting and destroying rare or antiquarian books across the industry.

    Anthropic told The Guardian that its acquisition programmes do not buy and destroy rare or antiquarian books. The identities and intentions of buyers behind many of the unusual bulk orders remain unclear.

    So the responsible conclusion is:

    Mass destructive scanning is documented. The broader destruction of genuinely rare books is a serious concern, but it has not been established at the same level of certainty.

    That nuance does not make the story unimportant. It makes the real story more credible.

    Why would an AI company want physical books?

    Because the open web is no longer enough.

    Modern AI models need enormous quantities of high-quality language. Books are especially valuable because they contain edited, structured, long-form thinking—something the internet does not always provide.

    Physical books also offer three advantages.

    1. They contain material that may not exist online

    Many older, specialist and out-of-print works were never turned into commercial ebooks. Their pages hold information that is effectively invisible to internet-scale data collection.

    2. Older books contain less AI-generated material

    As AI-generated text spreads across the web, training future systems on indiscriminate online data risks feeding models content produced by other models. Pre-generative-AI books are attractive because their human origin is easier to establish.

    3. Buying a physical copy can create a cleaner legal position

    The Anthropic ruling shows why acquisition method matters. Buying a copy, destroying it and keeping one internal digital replacement presented a stronger fair-use argument than downloading an unauthorised digital copy.

    In other words, this was not simply a knowledge project. It was also a data-sourcing and legal-risk strategy.

    The uncomfortable business lesson: your content is an input

    Most businesses still think about AI tools as products they consume.

    You pay for a chatbot, connect an API or add an AI assistant to your workflow. It feels like a normal software relationship: the vendor provides the tool, and you use it.

    But AI platforms are also built around inputs. They need language, images, behaviour, feedback and context. Your business may be a customer on one side of that system and a source of valuable data on the other.

    That does not mean every AI provider trains on every prompt or secretly takes every file. Policies, contracts and product settings differ. Enterprise and API offerings often include stronger data controls than free consumer tools.

    The point is simpler: never assume your content is protected merely because you created it or because it sits inside a tool you pay for. Protection comes from clear terms, technical controls and deliberate choices.

    What this means for ecommerce and marketing teams

    For an ecommerce business, “content” is not just blog posts.

    It includes product descriptions, photography, customer reviews, campaign concepts, brand voice, internal merchandising rules, conversion experiments, support tickets and pricing logic. Individually, these assets may look ordinary. Together, they describe how your company competes.

    If teams paste that material into AI tools without checking the terms, they may expose far more than a few paragraphs of copy.

    Consider four common situations:

    A marketer uploads next quarter’s campaign plan

    The document may contain unreleased offers, audience insights, budgets and positioning. The risk is not only copyright. It is confidentiality.

    A product team feeds an entire catalogue into a writing tool

    Generated descriptions may save time, but the input also reveals assortment strategy, attributes and product data. Who can retain it, and for how long?

    Customer service uses public AI tools to rewrite tickets

    Those tickets may contain names, addresses, order details or health information. Now the issue includes privacy and GDPR—not just content ownership.

    A creator builds a brand on a third-party model

    If the model, price, policy or output quality changes, the creator’s workflow can break overnight. Dependence becomes a platform risk.

    Five practical actions businesses should take now

    You do not need to stop using AI. You need to stop using it casually.

    1. Classify information before it enters an AI tool

    Create three simple categories: public, internal and restricted. Public material may be acceptable in approved tools. Internal content needs controls. Restricted data—such as personal information, credentials, contracts and unreleased financials—should not enter an unapproved system.

    2. Read the terms that matter

    Check whether the provider may retain inputs, use them to improve models, allow human review or share them with subprocessors. Confirm whether training is disabled by default, optional or unavailable for your plan.

    Do not let “enterprise-grade” function as a substitute for reading the contract.

    3. Keep an original source of truth

    Store product copy, research, images, prompts and campaign assets in systems you control. AI output should enter your workflow; your workflow should not live entirely inside one AI platform.

    4. Preserve human provenance

    Keep drafts, timestamps, licences and approval records for important creative assets. This helps demonstrate where work came from, what a human contributed and which material you had permission to use.

    5. Avoid single-model dependence

    Build processes around tasks and standards rather than one vendor’s interface. Where practical, keep prompts portable, retain exports and test a backup provider. The goal is not to switch tools every week. It is to maintain leverage.

    Legal does not automatically mean ethical—or wise

    The court’s decision addressed specific copyright questions under US law. It did not settle every ethical question raised by destroying physical books, nor did it create a universal rule for every AI model, dataset or country.

    A purchased mass-market paperback is not the same thing as a fragile edition with annotations, a distinctive binding or historical provenance. A digital text can preserve words while losing the object’s physical evidence.

    The environmental picture is complicated too. Recycling the paper is better than sending it to landfill, but buying, transporting, cutting and scanning millions of books still consumes material and energy. A company can follow a legally defensible process without proving it chose the most responsible one.

    For businesses, that distinction is essential. Compliance asks, “Are we allowed to do this?” Trust asks, “Will customers, creators and partners believe this is fair?”

    The strongest brands need an answer to both.

    The real story is not about paper

    Physical books make this issue visible because we understand what is being lost. We can picture the blade cutting through the binding. We can see the pages becoming data.

    Digital extraction is easier to ignore. A website can be scraped without an empty shelf. A creator’s style can be absorbed without a damaged cover. A customer conversation can become a data point without anyone hearing the paper shredder.

    That is why this story matters.

    AI is not magic floating above the economy. It is infrastructure built from human work: books, art, code, conversations, decisions and data. Businesses benefiting from these systems should ask where those inputs came from—and apply the same scrutiny to where their own information goes.

    Use AI. Experiment with it. Build with it.

    But do not confuse convenience with control.

    Frequently asked questions

    Are AI companies really destroying physical books?

    Yes, in at least one well-documented case. Court records confirm that Anthropic bought and destructively scanned millions of physical books, removing bindings or spines and recycling the remains after digitisation.

    Are AI companies destroying rare books?

    There are credible reports of unusual purchases involving obscure, old and out-of-print books, and booksellers suspect AI-related buyers. However, systematic destruction of genuinely rare or antiquarian books has not been conclusively established. Anthropic denies buying and destroying rare or antiquarian books through its acquisition programmes.

    Was Anthropic’s scanning ruled legal?

    In June 2025, a US federal judge held that converting lawfully purchased print books into internal digital replacements was fair use in the specific case. The same ruling did not excuse Anthropic’s acquisition and retention of pirated library copies.

    Can an AI company train on my business content?

    It depends on how the content is obtained, the provider’s terms, your product tier, applicable law and the settings or contract governing your account. Businesses should verify these conditions rather than assume all AI tools handle data in the same way.

    Should businesses stop using generative AI?

    No. Businesses should use approved tools, classify sensitive information, understand provider terms, preserve source files and avoid depending entirely on a single model or platform.

    Sources and further reading