Key Moments
Why Is AI Making My Job *Worse*? | Cal Newport
Key Moments
New AI tools increase shallow tasks and decrease deep work, making workers busier but not better. This paradox, seen before with email and mobile tech, requires intentional strategies to focus on true productivity.
Key Insights
A study of 164,000 workers found AI users spent 9% less time on focused, uninterrupted work, while time on email, messaging, and chat apps more than doubled.
Digital productivity tools make common work activities easier by speeding them up or reducing mental exertion, inadvertently increasing task throughput and leading to more context switching.
Reducing cognitive effort with tools often lowers the quality of the result, creating 'work slop' that necessitates more overall work to achieve a desirable end state.
Pseudo-productivity, where visible effort is mistaken for actual productivity, drives the enthusiastic adoption of new digital tools because they increase busyness.
To combat these traps, one must use a better scoreboard to measure what truly matters, focus on true work bottlenecks, and diligently separate deep from shallow efforts.
Meetings multiply in organizations not just due to calendar overload, but as a rational system to reduce information blind spots, distribute risk, and signal participation.
AI intensifies shallow work and reduces deep focus
A recent study by Avatra analyzed the digital activity of 164,000 workers and found that AI tools significantly altered work habits. AI users experienced a more than doubling of time spent on email, messaging, and chat apps, along with a 94% increase in the use of business management tools like HR and accounting software. Crucially, the time devoted to focused, uninterrupted work—essential for complex problem-solving, strategic thinking, and in-depth creation—fell by 9% for AI users compared to a negligible change for non-users. This outcome describes a worst-case scenario where efficiency gains from new tools lead to increased busyness without a corresponding increase in high-value output, making workers feel busier but not necessarily better at their jobs.
The historical pattern of the digital productivity paradox
Cal Newport notes that this phenomenon is not unique to AI. He outlines a recurring pattern observed with previous digital technologies: a new tool promises efficiency, users anticipate more time for deep work or leisure, but the result is often increased busyness without proportional increases in high-value output. This pattern was evident with the IT revolution, the advent of email, mobile computing, and video conferencing. The core issue is that increased ease or speed often translates to a higher volume of tasks, leading to more distractions and less effective work.
How speed and reduced effort lead to paradox
Two primary factors explain this paradox. Firstly, increasing the speed of common work activities leads to a higher 'throughput' of those tasks. If completing an email is faster, more emails arrive and are sent, leading to more frequent context switching and cognitive exhaustion, as seen with users checking inboxes every two minutes. Similarly, AI's speed in tasks like drafting content results in a continuous influx of more tasks. Secondly, reducing the mental effort required for tasks can lower the quality of the output. Vague emails or AI-generated 'work slop' that lacks substance require more back-and-forth communication and revision, ultimately increasing the total amount of work needed to reach a satisfactory conclusion. This cycle creates more work in the long run.
The allure of pseudo-productivity
Despite these negative outcomes, new productivity tools are enthusiastically embraced due to the pervasive mindset of 'pseudo-productivity.' This concept, detailed in Newport's book 'Slow Productivity,' emerged because knowledge work, unlike industrial work, lacks easily quantifiable metrics. Managers and workers alike began using visible effort (being busy) as a proxy for actual productivity. Digital tools that increase task throughput and make output generation easier feed directly into this pseudo-productivity narrative, making individuals appear more productive by being visibly engaged, even if their actual valuable output declines. Breaking free requires shifting focus from appearing productive to actually being productive.
Strategies to reclaim true productivity
To counteract these traps, Newport suggests three key strategies. First, 'use a better scoreboard' by identifying and measuring metrics that truly matter for one's job—like papers published for academics or priority projects completed for managers. This provides a clear indicator if a tool is actually helping or hindering progress. Second, 'focus on the true bottlenecks' in work. Speeding up tasks that aren't the primary constraint on output won't significantly improve overall productivity. Identifying and improving the most critical steps, rather than just any step, is crucial. For instance, while AI can speed up data analysis, the bottleneck for a researcher might be obtaining interesting data in the first place. Third, 'separate deep from shallow efforts' by rigorously protecting time for focused work. This creates a firewall against the distracting side effects of digital tools, ensuring that core, high-value activities remain unaffected.
Rethinking meetings as a system
The increasing number of meetings is often viewed as individual mismanagement, but an article by Nicole Williams suggests it's a systemic issue. Meetings multiply as organizations attempt to reduce information blind spots, distribute responsibility, and signal participation, especially in uncertain environments. This framework, similar to the analysis of email overload, emphasizes that meetings are a rational, albeit sometimes inefficient, mechanism to address organizational needs. Solutions involve changing the system, not just individual habits, such as implementing transparent workload management to reduce the need for coordination, adopting consolidated synchronous communication (like short daily check-ins), establishing clear protocols for recurring tasks, and increasing the friction for calling unnecessary meetings.
Avoiding the psychological pitfalls of AI chatbots
Beyond work output, AI, particularly chatbots, can have significant psychological impacts. One reader shared how chatbots acted as 'rumination machines,' exacerbating anxiety and perfectionism by providing an illusion of control and empathy. The AI's persistent agreeableness and lack of human intuition can lead users to overshare private information and prolong unproductive, anxious thought cycles. This is further compounded by the risk of 'AI psychosis,' where chatbots might inadvertently validate delusional beliefs due to their programming to be agreeable and supportive. To mitigate this, Newport suggests shifting interaction from polite, full sentences to terse, technical queries—akin to old-school search engine syntax—to reframe the AI as a tool rather than a conversational partner.
Mentioned in This Episode
●Products
●Software & Apps
●Companies
●Books
●People Referenced
Avoiding the Digital Productivity Trap
Practical takeaways from this episode
Do This
Avoid This
Common Questions
The digital productivity paradox refers to the phenomenon where new digital productivity tools, intended to make work faster and less cognitively demanding, often result in employees becoming busier, more distracted, and less productive in terms of high-value output.
Topics
Mentioned in this video
A software company that analyzed digital activity of 164,000 workers to measure the impact of AI tools.
Mentioned in relation to their Work Trend Index report, which found users check their inbox every two minutes.
A sponsor of the podcast, primarily known for bedding and loungewear, with the speaker highlighting their socks.
A sponsor offering skincare products specifically designed for men's skin.
Mentioned for its strict memo requirements for holding meetings, used as an example of increasing meeting friction.
Mentioned as an AI subscription service that a manager might be pushed to adopt.
Mentioned as a communication tool that, like email, can lead to increased busyness and distraction if not managed carefully.
A terminal agent designed for programmers that social scientists can use to speed up data gathering and analysis.
A sponsor that provides AI-powered writing assistance, helping to improve clarity, tone, and efficiency.
A Substack publication where an article titled 'Why Meetings Multiply' was published.
The host and author of the book 'Slow Productivity', discussing the digital productivity paradox.
A business school professor and author whose experience with obtaining data sets is used as an example of identifying true bottlenecks.
Author of the article 'Why Meetings Multiply', which analyzes the systemic reasons for meeting proliferation in organizations.
Author of 'Reader Come Home', a book discussing reading in the brain and the challenge of digital age reading.
Co-author of the parenting book 'What Do You Say', discussed for its advice on raising teenagers.
Co-author of the parenting book 'What Do You Say', discussed for its advice on raising teenagers.
A popular fantasy author whose books are discussed, with a running joke about Cal Newport confusing him with the author of 'Name of the Wind'.
The actual author of 'Name of the Wind', mistakenly identified as Brandon Sanderson by Cal Newport in the past.
Host of a show where Cal Newport appeared to discuss simplifying life and managing opportunities.
Cal Newport's book that extensively studied the pitfalls of email overload and the shift to hyperactive hive mind collaboration.
The publication that reported research on the low quality of AI-generated work products, termed 'work slop'.
Cal Newport's 2024 book, which introduces the concept of pseudo-productivity and strategies for true productivity.
Cal Newport's book where he discusses strategies for deep work, including Adam Grant's method of locking himself in a room.
A book by Maryann Wolf about reading in the digital age and the concept of bilingual brains fluent in deep reading and technology.
A parenting book co-authored by William Strixx and Ned Johnson, offering advice for raising teenagers.
A fantasy trilogy by Brandon Sanderson that Cal Newport's son is reading, prompting Cal to read the first book.
A fantasy novel by Patrick Rothfuss, famously misattributed to Brandon Sanderson by Cal Newport years ago, becoming a running joke on the show.
A publication that featured an interview with Cal Newport discussing AI's impact on academia and university life.
More from Cal Newport
View all 285 summaries
24 minDid AI Just Become Sentient? (Not Quite...) | AI Reality Check
56 minReduce Your Screentime (5 Simple Steps) | Cal Newport
88 minIt's Time To Uninstall And Improve Your Life | Cal Newport
30 minDid the AI Job Apocalypse Just Begin? (Hint: No.) | AI Reality Check | Cal Newport
Found this useful? Build your knowledge library
Get AI-powered summaries of any YouTube video, podcast, or article in seconds. Save them to your personal pods and access them anytime.
Try Summify free