Amanda Kavner PhD ๐Ÿง ๐Ÿค–๐Ÿพ (@DrKavner) header
Amanda Kavner PhD ๐Ÿง ๐Ÿค–๐Ÿพ (@DrKavner) profile picture

PhD in Cognitive Neuroscience ๐Ÿง , Futurist ๐Ÿ”ฎ, SUNY professor ๐ŸŽ“, International Speaker ๐Ÿ—ฃ๏ธ, #cogsci/#Web3/#AI Educational Consultant/Advisor

Long Island, NY
Joined January 2014
Why AI Is Making Knowledge Workers Busier Without Making Them Sharper Most people who use AI at work can name the hours it saves, and far fewer can point to anything that time turned into, which is strange when you consider how much of the pitch for these tools rests on the promise that freeing up time would give us room to think more carefully about [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t3656361b
1
32
When Your AI Agent Negotiates Against Theirs, Who Is Actually at the Table? Sometime in the next few years, a copy of you that never sleeps is going to haggle over your cable bill with a stand-in for the provider that never tires of saying no, and neither of you will be in the room when the deal closes. That sounds like a convenience story, and for the [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t4581a91c
1
46
The Alignment Tax Inside Organizations The most expensive thing in most organizations isnโ€™t software licenses or headcount. Itโ€™s the cognitive drag created when people donโ€™t share the same definition of what theyโ€™re trying to do. This is the alignment tax: the hidden cost of ambiguity, mismatched goals, and vague language inside teams. Every hour spent relitigating decisions already made, every [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/ta099a838
25
How to Use AI for Sensemaking When the Situation Is Unclear The problem isnโ€™t that people donโ€™t have enough information. Itโ€™s that they have too much of it, and no coherent way to tell what matters. When a problem is genuinely messy, the usual approach is to dump everything into an AI chat and ask for a summary. Thatโ€™s the worst thing you can do. Summarization [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tc7edf249
41
The Compound Cost of Small AI Conveniences Most people think about AIโ€™s effect on their skills as a future problem. Something that might happen if theyโ€™re not careful, something to watch for. The research suggests itโ€™s already happening, quietly, in work they did last week. The mechanism is straightforward: skills that donโ€™t get exercised degrade, and AI removes the exercise without removing [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tf6a05e2c
30
The Hidden Cost of AI Speed Is Premature Closure There is a version of using AI that feels like efficient thinking and is actually the absence of it. The output is clean, the answer arrives fast, and the work gets done. What doesnโ€™t happen is the part where you sit with the uncertainty long enough to actually understand it. This is what researchers are [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tb53b5c57
40
What Happens to Your Writing Voice When You Use AI Every Day Your writing voice is a cognitive fingerprint. Itโ€™s built from the specific way your mind moves through a problem, the rhythm you default to under pressure, the metaphors that surface when youโ€™re not editing yourself, the things you notice that other people skip over. It takes years to develop and it degrades quietly when you [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t1a7a74dc
42
Cognitive Sovereignty in Teams: When Group AI Use Erodes Collective Judgment When a team starts using the same AI tool for the same kinds of work, something predictable happens that most organizations never notice: their outputs start to resemble each other. Not because people are collaborating more, but because theyโ€™re all drawing from the same underlying model, trained on the same data, with the same tendencies [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t15f1c05f
40
The Autonomy Gradient: A Framework for Deciding How Much to Delegate to AI Most advice about using AI at work treats delegation as a binary: either you use the tool or you donโ€™t. That framing misses the question that actually matters, which isnโ€™t whether to delegate but how much, and how to decide that for each specific task given what it costs you cognitively if you get it [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tb67a7624
45
What Neuroscience Says About the Attention You Have Left at the End of the Day Most productivity advice treats the workday as a flat surface. Eight hours, more or less equivalent, to be allocated toward the highest-priority tasks. What the research actually shows is that the workday has a shape, and scheduling against it rather than with it is one of the most common and least-noticed sources of cognitive underperformance. [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tada9bdb9
49
The Slow Return: How to Rebuild Focus After Long-Term Fragmentation The question isnโ€™t whether fragmented attention is a problem. At this point, thatโ€™s well established. The question most people havenโ€™t answered is what comes next, specifically what it actually takes to rebuild the capacity for sustained focus once itโ€™s been eroded, and whether the erosion is even reversible at all. The honest answer is: yes, [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t20e12876
34
Why Boredom Is a Cognitive State Worth Protecting Boredom has a bad reputation it doesnโ€™t entirely deserve. It feels like a problem to solve, a gap to fill, a signal that something has gone wrong with how youโ€™re spending your time, and most of us are now equipped with a device that makes filling the gap instantaneous. The phone comes out before the [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t32d65311
1
82
The Difference Between Assisted Thinking and Replaced Thinking Thereโ€™s a version of AI assistance where you end up sharper than you started, and thereโ€™s a version where you end up dependent, and from the outside they can look almost identical for months before the difference shows up. The output quality stays roughly the same. The timeline improves. The deliverable lands. What changes, slowly [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t318a0e5e
1
1
54
What Cognitive Sovereignty Actually Looks Like in Practice Cognitive sovereignty gets used as a slogan more often than a standard. You hear it in AI policy discussions, in education debates, in op-eds about what humans should protect as AI systems get more capable. It tends to mean something like โ€œthe ability to think for yourself,โ€ which sounds clear until you try to apply [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tc96117c2
41
The Focus Tax: What Every Notification Actually Costs The average person receives somewhere between 65 and 80 push notifications per day. Most of them take less than five seconds to glance at. That sounds like a small tax on your attention, almost negligible, the kind of thing youโ€™d wave away if someone pointed it out. But the cost of a notification isnโ€™t measured [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t5118fa12
29
The Attention Debt: How Years of Fragmented Focus Change What Deep Work Feels Like Thereโ€™s a specific kind of tired that isnโ€™t sleep deprivation. Itโ€™s the feeling at the end of a day where youโ€™ve been busy for eight hours and done almost nothing that required more than a few minutes of unbroken thought. Youโ€™re exhausted without having gone anywhere. The work got done, technically, but the version of [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tf623aa28
43
Relearning Boredom: Why Unpleasant Is the Goal Nobody likes being bored. Thatโ€™s the point. The discomfort of boredom is not a signal that something has gone wrong. Itโ€™s the signal that your brain is transitioning into a state that does things focused work canโ€™t do, things like integrating experiences, generating unprompted connections, and producing the kind of thinking that feels like it [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t75265e99
44
What You Get When the Robot Lifts the Barbell: The Cognitive Cost of Trading Process for Product Thereโ€™s a version of AI assistance that makes you better, and a version that makes you dependent. From the outside, they look identical. The output quality is the same. The speed is the same. The experience of using the tool is the same. The difference only shows up when you put the tool down. The [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t2f49c601
1
41
The Attention Economy Didnโ€™t End. It Went Deeper. The original attention economy had a simple structure: platforms competed for your eyes, and you were the product. You scrolled, they sold your attention to advertisers, and the feedback loop optimized endlessly for engagement. It was extractive and manipulative, but it was at least legible. You could see the feed. You could feel the pull. [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/tb32176c3
45
The Dopamine Hijack Worked. The AI Takeover Might Not. Hereโ€™s Why Your Brain Resists Differently. Social media did something to us that took years to fully understand, and weโ€™re still living with the consequences. The dopamine hijack was real, it was deliberate, and it worked so well precisely because it exploited reward pathways that evolved for something else entirely, things like social connection, status, and belonging, and turned them into [โ€ฆ] ๐—š๐—ฒ๐˜ ๐˜๐—ต๐—ฒ ๐—”๐—œ ๐—”๐˜‚๐—ฑ๐—ถ๐˜ ๐—ณ๐—ผ๐—ฟ ๐˜†๐—ผ๐˜‚๐—ฟ ๐—ฏ๐˜‚๐˜€๐—ถ๐—ป๐—ฒ๐˜€๐˜€ ๐—ต๐—ฒ๐—ฟ๐—ฒ: rfr.bz/t309b9661
38