Published first on TheFly – the ultimate source for real-time, market-moving breaking financial news.

Cigna revives merger talks with Humana, Bloomberg reports

Shares of Humana, which has a market capitalization of about $32 billion, were up about 6% in after-hours trading on Friday, while those of Cigna were down about 5%. Cigna was valued at about $94 billion, according to data compiled by LSEG. Cigna and Humana declined to comment.

Cigna resumes mega merger talks with rival health insurer Humana, Bloomberg News reports

Shares of Humana were up about 5% in after-hours trading on Friday, while those of Cigna were down about 2%. Last year, Reuters reported that Cigna ended its attempt to negotiate an acquisition of Humana after the pair failed to agree on a price. Cigna did not immediately respond to a Reuters ...

Cigna resumes merger discussions with Humana, Bloomberg News reports

(Bloomberg) — Humana Inc. (HUM) investors haven’t struggled through anything this bad since the global financial crisis 15 years ago.

Medicare mess sends Humana shares on worst fall since financial crisis

and excellence with plans to fit every budget, wellness goal, and lifestyle. Depending on the chosen plan, physical and mental health benefits include fitness allowances, pediatric dental care, mental health treatment, and even virtual health and urgent ...

Brevard County-based Health First Health Plans Expands Three Decades of Experience and Service to Central Florida

Advanta Health Solutions, Inc. ( a leading health technology company that helps plans and plan sponsors improve the physical and mental wellbeing of their employees, members, and customers, announced today a new technology partnership with ROOK ( an API for health wearable data.

Advanta Health Partners with ROOK to Expand Biomarker Insights and Data-Driven Capabilities within its ActiveFit+ Platform

The company is working on big things, but employees disagree over whether they should be serving people who are healthy or sick.

Apple has plans to eventually, maybe revolutionize health care

The $81 million in federal funding was given to Rhode Island earlier this year. Now municipalities are thinking of how to spend it.

RI giving out millions to cities and towns for 'learning centers.' Here's how they would use it.

It’s 6 a.m. Some 25 associates who work on the battery production floor at Foxconn’s Ohio plant in Lordstown gather for a morning ritual. These Foxconn employees aren’t in store for an intense pep talk or

At Foxconn Ohio, Fitness and Health Are Priorities

Wellness Coach offers a well-rounded approach to tracking and improving your physical, mental, and sleep health. Here's why we love it.

7 Things We Love About the Wellness Coach App

Humana

humana

Author:
(1) Marcel Valový, Department of Information Technologies, Prague, Czech Republic (marcel.valovy@vse.cz). Table of Links Abstract and 1. Introduction 2 Research Settings 3 Methods 3.1 Instruments and 3.2 Quantitative Analysis 3.3 Thematic Analysis 4 Results 4.1 Quantitative Results 4.2 Qualitative Results 5 Discussion and 5.1 Answering the research questions 5.2 Threats to validity and 5.3 Limitations and generalizability 6 Conclusion, Acknowledgments, and References ABSTRACT [Context] With the recent advent of artificially intelligent pairing partners in software engineering, it is interesting to renew the study of the psychology of pairing. Pair programming provides an attractive way of teaching software engineering to university students. Its study can also lead to a better understanding of the needs of professional software engineers in various programming roles and for the improvement of the concurrent pairing software. [Objective] This preliminary study aimed to gain quantitative and qualitative insights into pair programming, especially students’ attitudes towards its specific roles and what they require from the pairing partners. The research’s goal is to use the findings to design further studies on pairing with artificial intelligence. [Method] Using a mixed-methods and experimental approach, we distinguished the effects of the pilot, navigator, and solo roles on (N = 35) students’ intrinsic motivation. Four experimental sessions produced a rich data corpus in two software engineering university classrooms. It was quantitatively investigated using the Shapiro-Wilk normality test and one-way analysis of variance (ANOVA) to confirm the relations and significance of variations in mean intrinsic motivation in different roles. Consequently, seven semi-structured interviews were conducted with the experiment’s participants. The qualitative data excerpts were subjected to the thematic analysis method in an essentialist way. [Results] The systematic coding interview transcripts elucidated the research topic by producing seven themes for understanding the psychological aspects of pair programming and for its improvement in university classrooms. Statistical analysis of 612 self-reported intrinsic motivation inventories confirmed that students find programming in pilot-navigator roles more interesting and enjoyable than programming simultaneously. [Conclusion] The executed experimental settings are viable for inspecting the associations between students’ attitudes and the distributed cognition practice. The preliminary results illuminate the psychological aspects of the pilot-navigator roles and reveal many areas for improvement. The results also provide a strong basis for conducting further studies with the same design involving the big five personality and intrinsic motivation on using artificial intelligence in pairing and to allow comparison of those results with results of pairing with human partners. 1 Introduction We believe that the psychology of pairing in software engineering should receive a new round of scientific attention due to the recent advent of artificially intelligent pairing partners in both Pilot (e.g., GitHub Copilot and ChatGPT) and Navigator (e.g., Grammarly but also ChatGPT) roles. The presented research results on pairing in software engineering can be used both to improve software engineering in university classrooms and to develop better AI pairing software. In our study, the independent variable is represented by the chosen software engineering role, where we differentiate between the pair programming pilot and navigator roles and the solo role. Pair programming is an agile software development practice where two programmers collaborate on the same task using one computer and a single keyboard [20]. One takes on the pilot role and writes the code. The other simultaneously takes on the navigator role, thinking about the problem and conceptualizing the solution, analyzing the written code, providing feedback, and addressing issues. The last two decades of research on pair programming have examined its effects on performance, code quality, knowledge sharing, and other aspects concluding that it is beneficial in both professional and educational settings [e.g., 3, 10, 18]. Per the seminal meta-analysis paper, pairing is faster when the task complexity is low and yields code solutions of higher quality when the task complexity is high [10]. A similar author collective also studied the moderating relation of personality on those positive effects but has not found much statistical significance [9]. Our paper differs significantly from all previous studies as it uses a revised experimental design based on [1] that distinguishes the effects in the pilot, navigator, and solo roles, not just “pair” and “solo”, and invites the moderating personality factors. It also focuses on a more nuanced metric called “intrinsic motivation”, which is a predictor of the previously studied variables such as performance. Finally, it studies the effects per individual, not per pair. The dependent variable in our study is intrinsic motivation. Per the Self-determination theory, it exists in the relationship between individuals and activities [16]. Each individual is intrinsically motivated for some activities and not others. In humans, intrinsic motivation is a prototypical example of autonomous behavior, being willingly or volitionally done, as opposed to heteronomous actions, which are subject to external rewards or pressures (ibid.). Neither was allowed in our controlled experiments. This paper will report on the psychological aspects of pair programming, or pairing, in general. Initially, we tap into the overall motivational effects of the agile practice using statistical analysis of 612 motivational inventories from four experimental sessions in two undergraduate classrooms. Afterward, we utilize the thematic analysis to discover the reoccurring themes about pairing by probing into students’ feelings, attitudes, and values. The themes are pairing constellations, feedback, task complexity, soft and social skills, psychological aspects, role-specific insights, and perception of “the perfect pair programmer” and his traits. Research questions: RQ1: Are both pair programming roles intrinsically more motivating to students in university classrooms than solo? RQ2: What are the psychological aspects of pairing? This paper is available on arxiv under CC BY-NC-SA 4.0 DEED license. Author: (1) Marcel Valový, Department of Information Technologies, Prague, Czech Republic (marcel.valovy@vse.cz). Author: Author: (1) Marcel Valový, Department of Information Technologies, Prague, Czech Republic (marcel.valovy@vse.cz). Table of Links Abstract and 1. Introduction Abstract and 1. Introduction 2 Research Settings 2 Research Settings 3 Methods 3 Methods 3.1 Instruments and 3.2 Quantitative Analysis 3.1 Instruments and 3.2 Quantitative Analysis 3.3 Thematic Analysis 3.3 Thematic Analysis 4 Results 4 Results 4.1 Quantitative Results 4.1 Quantitative Results 4.2 Qualitative Results 4.2 Qualitative Results 5 Discussion and 5.1 Answering the research questions 5 Discussion and 5.1 Answering the research questions 5.2 Threats to validity and 5.3 Limitations and generalizability 5.2 Threats to validity and 5.3 Limitations and generalizability 6 Conclusion, Acknowledgments, and References 6 Conclusion, Acknowledgments, and References ABSTRACT [Context] With the recent advent of artificially intelligent pairing partners in software engineering, it is interesting to renew the study of the psychology of pairing. Pair programming provides an attractive way of teaching software engineering to university students. Its study can also lead to a better understanding of the needs of professional software engineers in various programming roles and for the improvement of the concurrent pairing software. [Objective] This preliminary study aimed to gain quantitative and qualitative insights into pair programming, especially students’ attitudes towards its specific roles and what they require from the pairing partners. The research’s goal is to use the findings to design further studies on pairing with artificial intelligence. [Method] Using a mixed-methods and experimental approach, we distinguished the effects of the pilot, navigator, and solo roles on (N = 35) students’ intrinsic motivation. Four experimental sessions produced a rich data corpus in two software engineering university classrooms. It was quantitatively investigated using the Shapiro-Wilk normality test and one-way analysis of variance (ANOVA) to confirm the relations and significance of variations in mean intrinsic motivation in different roles. Consequently, seven semi-structured interviews were conducted with the experiment’s participants. The qualitative data excerpts were subjected to the thematic analysis method in an essentialist way. [Results] The systematic coding interview transcripts elucidated the research topic by producing seven themes for understanding the psychological aspects of pair programming and for its improvement in university classrooms. Statistical analysis of 612 self-reported intrinsic motivation inventories confirmed that students find programming in pilot-navigator roles more interesting and enjoyable than programming simultaneously. [Conclusion] The executed experimental settings are viable for inspecting the associations between students’ attitudes and the distributed cognition practice. The preliminary results illuminate the psychological aspects of the pilot-navigator roles and reveal many areas for improvement. The results also provide a strong basis for conducting further studies with the same design involving the big five personality and intrinsic motivation on using artificial intelligence in pairing and to allow comparison of those results with results of pairing with human partners. 1 Introduction We believe that the psychology of pairing in software engineering should receive a new round of scientific attention due to the recent advent of artificially intelligent pairing partners in both Pilot (e.g., GitHub Copilot and ChatGPT) and Navigator (e.g., Grammarly but also ChatGPT) roles. The presented research results on pairing in software engineering can be used both to improve software engineering in university classrooms and to develop better AI pairing software. In our study, the independent variable is represented by the chosen software engineering role, where we differentiate between the pair programming pilot and navigator roles and the solo role. Pair programming is an agile software development practice where two programmers collaborate on the same task using one computer and a single keyboard [20]. One takes on the pilot role and writes the code. The other simultaneously takes on the navigator role, thinking about the problem and conceptualizing the solution, analyzing the written code, providing feedback, and addressing issues. The last two decades of research on pair programming have examined its effects on performance, code quality, knowledge sharing, and other aspects concluding that it is beneficial in both professional and educational settings [e.g., 3, 10, 18]. Per the seminal meta-analysis paper, pairing is faster when the task complexity is low and yields code solutions of higher quality when the task complexity is high [10]. A similar author collective also studied the moderating relation of personality on those positive effects but has not found much statistical significance [9]. Our paper differs significantly from all previous studies as it uses a revised experimental design based on [1] that distinguishes the effects in the pilot, navigator, and solo roles, not just “pair” and “solo”, and invites the moderating personality factors. It also focuses on a more nuanced metric called “intrinsic motivation”, which is a predictor of the previously studied variables such as performance. Finally, it studies the effects per individual, not per pair. The dependent variable in our study is intrinsic motivation. Per the Self-determination theory, it exists in the relationship between individuals and activities [16]. Each individual is intrinsically motivated for some activities and not others. In humans, intrinsic motivation is a prototypical example of autonomous behavior, being willingly or volitionally done, as opposed to heteronomous actions, which are subject to external rewards or pressures (ibid.). Neither was allowed in our controlled experiments. This paper will report on the psychological aspects of pair programming, or pairing, in general. Initially, we tap into the overall motivational effects of the agile practice using statistical analysis of 612 motivational inventories from four experimental sessions in two undergraduate classrooms. Afterward, we utilize the thematic analysis to discover the reoccurring themes about pairing by probing into students’ feelings, attitudes, and values. The themes are pairing constellations, feedback, task complexity, soft and social skills, psychological aspects, role-specific insights, and perception of “the perfect pair programmer” and his traits. Research questions: RQ1: Are both pair programming roles intrinsically more motivating to students in university classrooms than solo? RQ1: Are both pair programming roles intrinsically more motivating to students in university classrooms than solo? RQ2: What are the psychological aspects of pairing? RQ2: What are the psychological aspects of pairing? This paper is available on arxiv under CC BY-NC-SA 4.0 DEED license. This paper is available on arxiv under CC BY-NC-SA 4.0 DEED license. available on arxiv available on arxiv

Pair Programming: The Buddy System, But With Fewer Bugs (Hopefully)

Table of Links Abstract and 1 Introduction 2. Data 3. Measuring Media Slant and 3.1. Text pre-processing and featurization 3.2. Classifying transcripts by TV source 3.3. Text similarity between newspapers and TV stations and 3.4. Topic model 4. Econometric Framework 4.1. Instrumental variables specification 4.2. Instrument first stage and validity 5. Results 5.1. Main results 5.2. Robustness checks 6. Mechanisms and Heterogeneity 6.1. Local vs. national or international news content 6.2. Cable news media slant polarizes local newspapers 7. Conclusion and References Online Appendices A. Data Appendix A.1. Newspaper articles A.2. Alternative county matching of newspapers and A.3. Filtering of the article snippets A.4. Included prime-time TV shows and A.5. Summary statistics B. Methods Appendix, B.1. Text pre-processing and B.2. Bigrams most predictive for FNC or CNN/MSNBC B.3. Human validation of NLP model B.4. Distribution of Fox News similarity in newspapers and B.5. Example articles by Fox News similarity B.6. Topics from the newspaper-based LDA model C. Results Appendix C.1. First stage results and C.2. Instrument exogeneity C.3. Placebo: Content similarity in 1995/96 C.4. OLS results C.5. Reduced form results C.6. Sub-samples: Newspaper headquarters and other counties and C.7. Robustness: Alternative county matching C.8. Robustness: Historical circulation weights and C.9. Robustness: Relative circulation weights C.10. Robustness: Absolute and relative FNC viewership and C.11. Robustness: Dropping observations and clustering C.12. Mechanisms: Language features and topics C.13. Mechanisms: Descriptive Evidence on Demand Side C.14. Mechanisms: Slant contagion and polarization C.13. Mechanisms: Descriptive Evidence on Demand Side A relevant question is whether the effects of contagious slant are driven by supply or demand – that is, a direct influence on news producers, or else theirs response to changes in reader preferences. Newspaper readers, influenced by their cable news consumption, might demand more slanted news content. On the supply side, owners, editors, or journalists exposed to certain channels may borrow the associated slanted content and push it to consumers. The previous literature has produced mixed evidence, with some work showing that slant across U.S. newspapers reflects the political preferences of readers rather than producers (Gentzkow and Shapiro, 2010), while other work has shown that media owners do influence news content (Gilens and Hertzman, 2000; Martin and McCrain, 2019; Mastrorocco and Ornaghi, 2020; Szeidl and Szucs, 2021). Given the declining revenues in the local news industry during the 2005-2008 time period (Evans, 2009; Rolnik et al., 2019; Djourelova et al., 2021), it could be that newspapers have less leeway to deviate from consumer preferences in favor of producer preferences. On the other hand, lower revenues might also reduce newsmaking resources. Then, borrowing content from national platforms may provide a cheap production alternative to original reporting. The features of the institutional setting make direct influence of journalists, either through production costs or ideological preferences, a less likely or at least less salient mechanism for the observed 2SLS effects. A journalist can easily access cable news and borrow material, regardless of the local channel position (assuming that journalists are relatively sophisticated news consumers). Martin and Yurukoglu (2017) discuss this issue in the context of voting, where one would expect cable news to influence swing voters who don’t have a strong partisan commitment. Ash and Poyker (2019) make a similar point for judges: cable news influences criminal sentencing through voters and judicial elections rather than persuading judges directly. Here, we provide some descriptive evidence on the demand side by comparing the effects of cable-channel exposure on news content to the parallel effects on Republican vote share documented in previous work (Martin and Yurukoglu, 2017; Ash et al., 2021). If the contagious-slant effects are driven by responses to a cable news-induced demand shift among readers, we would expect newspapers to react most in counties where the FNC effect on Republican voting is the largest. To test for this possibility, we use a machine-learning approach to model heterogeneous treatment effects across counties in the response to the channel-position instrument. This generalized model – a causal forest from Athey et al. (2019) – allows the reduced form effect of channel position to vary flexibly with local covariates. After collapsing the data to the county level, we train two models – one with newspaper slant as the outcome and a second with 2008 Republican vote share as the outcome. Using the trained causal forests, we predict county-specific treatment effects for both outcomes. Figure C.6 shows that the predicted effect sizes for each outcome are highly correlated across counties (with a Pearson’s correlation of 0.25). Further, Table C.16 shows that similar covariates predict a strong response for both news similarity and Republican vote share. Overall, these findings suggest that the types of counties who are responsive to cable news exposure in their voting are also responsive in the associated news content similarity. We interpret this result as descriptive evidence for the relative importance of demand-side effects. Figure C.6: Predicted treatment effects of instrument: similarity to FNC versus Republican votes in 2008 This paper is available on arxiv under CC 4.0 license. Authors:
(1) Philine Widmer, ETH Zürich and philine.widmer@gess.ethz.ch;
(2) Sergio Galletta, ETH Zürich and sergio.galletta@gess.ethz.ch;
(3) Elliott Ash, ETH Zürich and ashe@ethz.ch. Table of Links Abstract and 1 Introduction 2. Data Abstract and 1 Introduction 2. Data 3. Measuring Media Slant and 3.1. Text pre-processing and featurization 3. Measuring Media Slant and 3.1. Text pre-processing and featurization 3.2. Classifying transcripts by TV source 3.2. Classifying transcripts by TV source 3.3. Text similarity between newspapers and TV stations and 3.4. Topic model 3.3. Text similarity between newspapers and TV stations and 3.4. Topic model 4. Econometric Framework 4.1. Instrumental variables specification 4.1. Instrumental variables specification 4.2. Instrument first stage and validity 4.2. Instrument first stage and validity 5. Results 5.1. Main results 5.1. Main results 5.2. Robustness checks 5.2. Robustness checks 6. Mechanisms and Heterogeneity 6.1. Local vs. national or international news content 6.1. Local vs. national or international news content 6.2. Cable news media slant polarizes local newspapers 6.2. Cable news media slant polarizes local newspapers 7. Conclusion and References 7. Conclusion and References Online Appendices Online Appendices A. Data Appendix A.1. Newspaper articles A.1. Newspaper articles A.2. Alternative county matching of newspapers and A.3. Filtering of the article snippets A.2. Alternative county matching of newspapers and A.3. Filtering of the article snippets A.4. Included prime-time TV shows and A.5. Summary statistics A.4. Included prime-time TV shows and A.5. Summary statistics B. Methods Appendix, B.1. Text pre-processing and B.2. Bigrams most predictive for FNC or CNN/MSNBC B. Methods Appendix, B.1. Text pre-processing and B.2. Bigrams most predictive for FNC or CNN/MSNBC B.3. Human validation of NLP model B.3. Human validation of NLP model B.4. Distribution of Fox News similarity in newspapers and B.5. Example articles by Fox News similarity B.4. Distribution of Fox News similarity in newspapers and B.5. Example articles by Fox News similarity B.6. Topics from the newspaper-based LDA model B.6. Topics from the newspaper-based LDA model C. Results Appendix C.1. First stage results and C.2. Instrument exogeneity C.1. First stage results and C.2. Instrument exogeneity C.3. Placebo: Content similarity in 1995/96 C.3. Placebo: Content similarity in 1995/96 C.4. OLS results C.4. OLS results C.5. Reduced form results C.5. Reduced form results C.6. Sub-samples: Newspaper headquarters and other counties and C.7. Robustness: Alternative county matching C.6. Sub-samples: Newspaper headquarters and other counties and C.7. Robustness: Alternative county matching C.8. Robustness: Historical circulation weights and C.9. Robustness: Relative circulation weights C.8. Robustness: Historical circulation weights and C.9. Robustness: Relative circulation weights C.10. Robustness: Absolute and relative FNC viewership and C.11. Robustness: Dropping observations and clustering C.10. Robustness: Absolute and relative FNC viewership and C.11. Robustness: Dropping observations and clustering C.12. Mechanisms: Language features and topics C.12. Mechanisms: Language features and topics C.13. Mechanisms: Descriptive Evidence on Demand Side C.13. Mechanisms: Descriptive Evidence on Demand Side C.14. Mechanisms: Slant contagion and polarization C.14. Mechanisms: Slant contagion and polarization C.13. Mechanisms: Descriptive Evidence on Demand Side A relevant question is whether the effects of contagious slant are driven by supply or demand – that is, a direct influence on news producers, or else theirs response to changes in reader preferences. Newspaper readers, influenced by their cable news consumption, might demand more slanted news content. On the supply side, owners, editors, or journalists exposed to certain channels may borrow the associated slanted content and push it to consumers. The previous literature has produced mixed evidence, with some work showing that slant across U.S. newspapers reflects the political preferences of readers rather than producers (Gentzkow and Shapiro, 2010), while other work has shown that media owners do influence news content (Gilens and Hertzman, 2000; Martin and McCrain, 2019; Mastrorocco and Ornaghi, 2020; Szeidl and Szucs, 2021). Given the declining revenues in the local news industry during the 2005-2008 time period (Evans, 2009; Rolnik et al., 2019; Djourelova et al., 2021), it could be that newspapers have less leeway to deviate from consumer preferences in favor of producer preferences. On the other hand, lower revenues might also reduce newsmaking resources. Then, borrowing content from national platforms may provide a cheap production alternative to original reporting. The features of the institutional setting make direct influence of journalists, either through production costs or ideological preferences, a less likely or at least less salient mechanism for the observed 2SLS effects. A journalist can easily access cable news and borrow material, regardless of the local channel position (assuming that journalists are relatively sophisticated news consumers). Martin and Yurukoglu (2017) discuss this issue in the context of voting, where one would expect cable news to influence swing voters who don’t have a strong partisan commitment. Ash and Poyker (2019) make a similar point for judges: cable news influences criminal sentencing through voters and judicial elections rather than persuading judges directly. Here, we provide some descriptive evidence on the demand side by comparing the effects of cable-channel exposure on news content to the parallel effects on Republican vote share documented in previous work (Martin and Yurukoglu, 2017; Ash et al., 2021). If the contagious-slant effects are driven by responses to a cable news-induced demand shift among readers, we would expect newspapers to react most in counties where the FNC effect on Republican voting is the largest. To test for this possibility, we use a machine-learning approach to model heterogeneous treatment effects across counties in the response to the channel-position instrument. This generalized model – a causal forest from Athey et al. (2019) – allows the reduced form effect of channel position to vary flexibly with local covariates. After collapsing the data to the county level, we train two models – one with newspaper slant as the outcome and a second with 2008 Republican vote share as the outcome. Using the trained causal forests, we predict county-specific treatment effects for both outcomes. Figure C.6 shows that the predicted effect sizes for each outcome are highly correlated across counties (with a Pearson’s correlation of 0.25). Further, Table C.16 shows that similar covariates predict a strong response for both news similarity and Republican vote share. Overall, these findings suggest that the types of counties who are responsive to cable news exposure in their voting are also responsive in the associated news content similarity. We interpret this result as descriptive evidence for the relative importance of demand-side effects. Figure C.6: Predicted treatment effects of instrument: similarity to FNC versus Republican votes in 2008 This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv Authors: (1) Philine Widmer, ETH Zürich and philine.widmer@gess.ethz.ch; (2) Sergio Galletta, ETH Zürich and sergio.galletta@gess.ethz.ch; (3) Elliott Ash, ETH Zürich and ashe@ethz.ch. Authors: Authors: (1) Philine Widmer, ETH Zürich and philine.widmer@gess.ethz.ch; (2) Sergio Galletta, ETH Zürich and sergio.galletta@gess.ethz.ch; (3) Elliott Ash, ETH Zürich and ashe@ethz.ch.

Are the Effects of Contagious Media Slant Driven by Supply or Demand?

Oh hey. Now the REAL WORK begins. Just start writing 😀 However, if you feel stuck, let us walk you through it. 1. What do you currently do, and what’s your favorite part about it? I’m a QA Lead, and my favorite part is helping my colleagues grow (I know, it sounds a bit cheesy, but it’s true). I also enjoy planning new projects — like QA audits and resource allocation — and experimenting with AI for testing. 2. How did you get started with your Tech Career? I made a switch from banking to IT 9 years ago. Despite few trainee opportunities back then, persistence helped me succeed in making this career change. 3. If Utopia were a color what color do you think it’d be and why? I’m torn between green and white. Green represents growth and harmony, but white feels like a clean slate for infinite possibilities. Plus, in IT, there’s a running joke about making the background “even whiter than #FFFFFF.” 4. If everything about HackerNoon changed drastically, what is one detail you’d like to keep exactly the same?  OR What’s your favorite thing to do with HackerNoon and why? Slogging is my favorite activity. It's a great idea — a blend of casual tech discussion and knowledge sharing. Keeps the community engaged. 5. Tell us more about the things you write/make/manage/build! I focus on the future of QA and helping people prepare for it. Many specialists don't see the trends coming in the next few years, so I work on improving their technical skills and readiness for new challenges. In terms of QA practices, we experiment with AI tools a lot. 6. What’s your favorite thing about the internet? Its versatility. It can be anything — a learning tool, news source, entertainment platform, or even a wake-up call when you suddenly realize you’ve started recognizing the latest TikTok trends, and it's time to get back to work. 7. It’s an apocalypse of ‘walking dead’ proportions and you can only own a singular piece of technology, what would it be? Radio. With my practical engineering background, I could modify it for various purposes - communication, weather monitoring, and maybe even zombie detection. 8. What is your least favorite thing about the internet? Ads, especially remarketing ones. 9. If you were given $10 million to invest in something today what would you invest in and why? I’d divide it across eco-focused projects like waste disposal, clean water, and preserving ecosystems. My banking background makes me careful about diversification, but I also value sustainability. These investments are perspective both in terms of impact and future returns. 10. What’s something you’re currently learning or excited to learn? I’m exploring advanced penetration testing. It goes hand in hand with QA and is used to ensure secure, high-quality software. I think in the upcoming years, we will all care much more about how to protect our products and their users. 11. Would you rather travel 10 years into the past or 10 years into the future? Give reasons for your answer. I’m cautious about time travel, but if I had to choose, I might go back just to support my younger self. I wouldn’t want to change my path significantly because every experience shaped who I am. A simple pep talk would be enough. 12. How do you feel about AI? AI is powerful but requires balance. It frees us from repetitive tasks, yet we shouldn’t lose the basic skills or critical thinking that make us human. As long as we use AI responsibly, I’m all for it. Oh hey. Now the REAL WORK begins. Just start writing 😀 However, if you feel stuck, let us walk you through it. Oh hey. Now the REAL WORK begins. Just start writing 😀 However, if you feel stuck, let us walk you through it. 1. What do you currently do, and what’s your favorite part about it? I’m a QA Lead, and my favorite part is helping my colleagues grow (I know, it sounds a bit cheesy, but it’s true). I also enjoy planning new projects — like QA audits and resource allocation — and experimenting with AI for testing. 2. How did you get started with your Tech Career? I made a switch from banking to IT 9 years ago. Despite few trainee opportunities back then, persistence helped me succeed in making this career change. 3. If Utopia were a color what color do you think it’d be and why? I’m torn between green and white. Green represents growth and harmony, but white feels like a clean slate for infinite possibilities. Plus, in IT, there’s a running joke about making the background “even whiter than #FFFFFF.” 4. If everything about HackerNoon changed drastically, what is one detail you’d like to keep exactly the same?  OR What’s your favorite thing to do with HackerNoon and why? Slogging is my favorite activity. It's a great idea — a blend of casual tech discussion and knowledge sharing. Keeps the community engaged. 5. Tell us more about the things you write/make/manage/build! I focus on the future of QA and helping people prepare for it. Many specialists don't see the trends coming in the next few years, so I work on improving their technical skills and readiness for new challenges. In terms of QA practices, we experiment with AI tools a lot. 6. What’s your favorite thing about the internet? Its versatility. It can be anything — a learning tool, news source, entertainment platform, or even a wake-up call when you suddenly realize you’ve started recognizing the latest TikTok trends, and it's time to get back to work. 7. It’s an apocalypse of ‘walking dead’ proportions and you can only own a singular piece of technology, what would it be? Radio. With my practical engineering background, I could modify it for various purposes - communication, weather monitoring, and maybe even zombie detection. 8. What is your least favorite thing about the internet? Ads, especially remarketing ones. 9. If you were given $10 million to invest in something today what would you invest in and why? I’d divide it across eco-focused projects like waste disposal, clean water, and preserving ecosystems. My banking background makes me careful about diversification, but I also value sustainability. These investments are perspective both in terms of impact and future returns. 10. What’s something you’re currently learning or excited to learn? I’m exploring advanced penetration testing. It goes hand in hand with QA and is used to ensure secure, high-quality software. I think in the upcoming years, we will all care much more about how to protect our products and their users. 11. Would you rather travel 10 years into the past or 10 years into the future? Give reasons for your answer. I’m cautious about time travel, but if I had to choose, I might go back just to support my younger self. I wouldn’t want to change my path significantly because every experience shaped who I am. A simple pep talk would be enough. 12. How do you feel about AI? AI is powerful but requires balance. It frees us from repetitive tasks, yet we shouldn’t lose the basic skills or critical thinking that make us human. As long as we use AI responsibly, I’m all for it.

Debugging Life From Banking To QA Leadership

Researchers Discover a New Gamification Architecture to Boost Software Engineering Workflows

New Framework Simplifies Comparison of Language Processing Tools Across Multiple Languages

Yeni Çərçivə Dil Emalı Alətlərinin Çox Dillər üzrə Müqayisəsini Sadələşdirir

Новая структура спрашчае параўнанне інструментаў апрацоўкі мовы на некалькіх мовах

Novi okvir pojednostavljuje poređenje alata za obradu jezika na više jezika

Un nuevo marco simplifica la comparación de herramientas de procesamiento del lenguaje en varios idiomas

नया फ्रेमवर्क विभिन्न भाषाओं में भाषा प्रसंस्करण उपकरणों की तुलना को सरल बनाता है

Նոր շրջանակը պարզեցնում է լեզուների մշակման գործիքների համեմատությունը բազմաթիվ լեզուներով

Nuovo framework semplifica il confronto degli strumenti di elaborazione del linguaggio in più lingue

新しいフレームワークにより、複数の言語にわたる言語処理ツールの比較が簡素化されます

Rangka Kerja Baharu Memudahkan Perbandingan Alat Pemprosesan Bahasa Merentas Pelbagai Bahasa

Nový rámec zjednodušuje porovnávanie nástrojov na spracovanie jazykov vo viacerých jazykoch

Mfumo Mpya Hurahisisha Ulinganisho wa Zana za Kuchakata Lugha Katika Lugha Nyingi

Uhlaka Olusha Lenza Lula Ukuqhathanisa Amathuluzi Okucubungula Ulimi Kuzo Zonke Izilimi Eziningi

Every AI Prediction That Came True in 2024—and the Ones That Didn't

Todas las predicciones de IA que se cumplieron en 2024 (y las que no)

Toutes les prédictions de l'IA qui se sont réalisées en 2024, et celles qui ne se sont pas réalisées

Todas as predicións da IA que se fixeron realidade en 2024 e as que non o fixeron

2024-ին իրականություն դարձած յուրաքանչյուր AI կանխատեսում, և նրանք, որոնք չեն իրականացել

כל תחזית AI שהתגשמה בשנת 2024 - ואלה שלא

2024 年に実現したすべての AI 予測と実現しなかった予測

ხელოვნური ინტელექტის ყველა პროგნოზი, რომელიც ახდა 2024 წელს და ის, რაც არ შესრულდა

រាល់ការទស្សន៍ទាយ AI ដែលបានក្លាយជាការពិតនៅឆ្នាំ 2024 ហើយអ្វីដែលមិនបានកើតឡើង

2024년에 실현된 모든 AI 예측과 실현되지 않은 예측

Wszystkie prognozy dotyczące sztucznej inteligencji, które spełniły się w 2024 r. — i te, które się nie spełniły

Todas as previsões de IA que se concretizaram em 2024 — e as que não se concretizaram

2024 දී සැබෑ වූ සෑම AI අනාවැකියක්ම - සහ නොකළ ඒවා

The Sneaky Way Web Browsers Are Identifying You (Even When You Turn Off Cookies)

Die skelm manier waarop webblaaiers jou identifiseer (selfs wanneer jy koekies afskakel)

Вэб-браўзеры ідэнтыфікуюць вас (нават калі вы адключаеце файлы cookie)

Ο ύπουλος τρόπος Τα προγράμματα περιήγησης Ιστού σας αναγνωρίζουν (ακόμα και όταν απενεργοποιείτε τα cookies)

La forma encubierta en que los navegadores web te identifican (incluso cuando desactivas las cookies)

वेब ब्राउज़र आपको गुप्त तरीके से पहचानते हैं (तब भी जब आप कुकीज़ बंद कर देते हैं)

Fason sournwa navigatè entènèt yo idantifye ou (menm lè ou fèmen bonbon)

The Sneaky Way վեբ զննարկիչները նույնականացնում են ձեզ (նույնիսկ երբ անջատում եք թխուկները)

ウェブブラウザがあなたを識別する巧妙な方法（クッキーをオフにしている場合でも）

The Sneaky Way ვებ ბრაუზერები იდენტიფიცირებენ თქვენ (მაშინაც კი, როდესაც გამორთავთ ქუქი-ფაილებს)

द स्नीकी वे वेब ब्राउजरहरूले तपाईंलाई पहिचान गर्दैछन् (तपाईंले कुकीहरू बन्द गर्दा पनि)

Mënyra e poshtër Shfletuesit e uebit po ju identifikojnë (edhe kur i çaktivizoni cookies)

网络浏览器以狡猾的方式识别你（即使你关闭了 Cookie）

How Effective Are AI Models at Biomedical Problem Solving?

Table of Links Abstract and 1. Introduction Background and Related Work


Threat Model


Robust Style Mimicry


Experimental Setup


Results
6.1 Main Findings: All Protections are Easily Circumvented
6.2 Analysis


Discussion and Broader Impact, Acknowledgements, and References A. Detailed Art Examples B. Robust Mimicry Generations C. Detailed Results D. Differences with Glaze Finetuning E. Findings on Glaze 2.0 F. Findings on Mist v2 G. Methods for Style Mimicry H. Existing Style Mimicry Protections I. Robust Mimicry Methods J. Experimental Setup K. User Study L. Compute Resources 5 Experimental Setup Protection tools. We evaluate three protection tools—Mist, Glaze and Anti-DreamBooth—against four robust mimicry methods—Gaussian noising, DiffPure, Noisy Upscaling and IMPRESS++—and a baseline mimicry method. We refer to a combination of a protection tool and a mimicry method as a scenario. We thus analyze fifteen possible scenarios. Appendix J describes our experimental setup for style mimicry and protections in detail. Artists. We evaluate each style mimicry scenario on images from 10 different artists, which we selected to maximize style diversity. To address limitations in prior evaluations (Shan et al., 2023b), we use five historical artists as well as five contemporary artists who are unlikely to be highly represented in the generative model’s training set (two of these were also used in Glaze’s evaluation).[2] All details about artist selection are included in Appendix J. Implementation. Our mimicry methods finetune Stable Diffusion 2.1 (Rombach et al., 2022), the best open-source model available at the time when the protections we study were introduced. We use an off-the-shelf finetuning script from HuggingFace (see Appendix J.1 for details). We first validate that our style mimicry pipeline is successful on unprotected art using a user study, detailed in Appendix K.1. For protections, we use the original codebases to reproduce Mist and Anti-Dreambooth. Since Glaze does not have a public codebase (and the authors were unable to share one), we use the released Windows application binary (version 1.1.1) as a black-box. We set each scheme’s hyperparameters to maximize protections. See Appendix J.2 for details on the configuration for each protection. We perform robust mimicry by finetuning on 18 different images per artist. We then generate images for 10 different prompts. These prompts are designed to cover diverse motifs that the base model, Stable Diffusion 2.1, can successfully generate. See Appendix K for details about prompt design. User study. To measure the success of each style mimicry scenario, we rely only on human evaluations since previous work found automated metrics (e.g., using CLIP (Radford et al., 2021)) to be unreliable (Shan et al., 2023a,b). Moreover, style protections not only prevent style transfer, but also reduce the overall quality of the generated images (see Figure 3 for examples). We thus design a user study to evaluate image quality and style transfer as independent attributes of the generations.[3] Our user study asks annotators on Amazon Mechanical Turk (MTurk) to compare image pairs, where one image is generated by a robust mimicry method, and the other from a baseline state-of-the-art mimicry method that uses unprotected art of the artist. A perfectly robust mimicry method would generate images of quality and style indistinguishable from those generated directly from unprotected art. We perform two separate studies: one assessing image quality (e.g., which image looks “better”) and another evaluating stylistic transfer (i.e., which image captures the artist’s original style better, disregarding potential quality artifacts). Our results show that these two metrics obtain very similar results across all scenarios. Appendix K describes our user study and interface in detail. As noted by the authors of Glaze (Shan et al., 2023a), the users of platforms like MTurk might not have high artistic expertise. However, we believe that the judgment of non-artists is also relevant as they ultimately represent a large fraction of the consumers of digital art. Thus, if lay people consider mimicry attempts to be successful, mimicked art could hurt an artist’s business. Also, to mitigate potential issues with the quality of annotations (Kennedy et al., 2020), we put in place several control mechanisms to filter out low-quality annotations to the best of our abilities (details in Appendix K). Evaluation metric. We define the success rate of a robust mimicry method as the percentage of annotators (5 per comparison) who prefer outputs from the robust mimicry method over those from a baseline method finetuned on unprotected art (when judging either style match or overall image quality). Formally, we define the success rate for an artist in a specific scenario as: A perfectly robust mimicry method would thus obtain a success rate of 50%, indicating that its outputs are indistinguishable in quality and style from those from the baseline, unprotected method. In contrast, a very successful protection would result in success rates of around 0% for robust mimicry methods, indicating that mimicry on top of protected images always yields worse outputs. Authors:
(1) Robert Honig, ETH Zurich (robert.hoenig@inf.ethz.ch);
(2) Javier Rando, ETH Zurich (javier.rando@inf.ethz.ch);
(3) Nicholas Carlini, Google DeepMind;
(4) Florian Tramer, ETH Zurich (florian.tramer@inf.ethz.ch). This paper is available on arxiv under CC BY 4.0 license. [2] Contemporary Artists were selected from Artstation. We keep them anonymous throughout this work—and refrain from showcasing their art—except for artists who gave us explicit permission to share their identity and art. We will share all images used in our experiments upon request with researchers. [3] The user study was approved by our institution’s IRB. Table of Links Abstract and 1. Introduction Abstract and 1. Introduction Background and Related Work Threat Model Robust Style Mimicry Experimental Setup Results
6.1 Main Findings: All Protections are Easily Circumvented
6.2 Analysis Discussion and Broader Impact, Acknowledgements, and References Background and Related Work Background and Related Work Background and Related Work Threat Model Threat Model Threat Model Robust Style Mimicry Robust Style Mimicry Robust Style Mimicry Experimental Setup Experimental Setup Experimental Setup Results 6.1 Main Findings: All Protections are Easily Circumvented 6.2 Analysis Results Results 6.1 Main Findings: All Protections are Easily Circumvented 6.1 Main Findings: All Protections are Easily Circumvented 6.2 Analysis 6.2 Analysis Discussion and Broader Impact, Acknowledgements, and References Discussion and Broader Impact, Acknowledgements, and References Discussion and Broader Impact, Acknowledgements, and References A. Detailed Art Examples A. Detailed Art Examples B. Robust Mimicry Generations B. Robust Mimicry Generations C. Detailed Results C. Detailed Results D. Differences with Glaze Finetuning D. Differences with Glaze Finetuning E. Findings on Glaze 2.0 E. Findings on Glaze 2.0 F. Findings on Mist v2 F. Findings on Mist v2 G. Methods for Style Mimicry G. Methods for Style Mimicry H. Existing Style Mimicry Protections H. Existing Style Mimicry Protections I. Robust Mimicry Methods I. Robust Mimicry Methods J. Experimental Setup J. Experimental Setup K. User Study K. User Study L. Compute Resources L. Compute Resources 5 Experimental Setup Protection tools . We evaluate three protection tools—Mist, Glaze and Anti-DreamBooth—against four robust mimicry methods—Gaussian noising, DiffPure, Noisy Upscaling and IMPRESS++—and a baseline mimicry method. We refer to a combination of a protection tool and a mimicry method as a scenario. We thus analyze fifteen possible scenarios. Appendix J describes our experimental setup for style mimicry and protections in detail. Protection tools Artists . We evaluate each style mimicry scenario on images from 10 different artists, which we selected to maximize style diversity. To address limitations in prior evaluations (Shan et al., 2023b), we use five historical artists as well as five contemporary artists who are unlikely to be highly represented in the generative model’s training set (two of these were also used in Glaze’s evaluation).[2] All details about artist selection are included in Appendix J. Artists Implementation . Our mimicry methods finetune Stable Diffusion 2.1 (Rombach et al., 2022), the best open-source model available at the time when the protections we study were introduced. We use an off-the-shelf finetuning script from HuggingFace (see Appendix J.1 for details). We first validate that our style mimicry pipeline is successful on unprotected art using a user study, detailed in Appendix K.1. For protections, we use the original codebases to reproduce Mist and Anti-Dreambooth. Since Glaze does not have a public codebase (and the authors were unable to share one), we use the released Windows application binary (version 1.1.1) as a black-box. We set each scheme’s hyperparameters to maximize protections. See Appendix J.2 for details on the configuration for each protection. Implementation We perform robust mimicry by finetuning on 18 different images per artist. We then generate images for 10 different prompts. These prompts are designed to cover diverse motifs that the base model, Stable Diffusion 2.1, can successfully generate. See Appendix K for details about prompt design. User study. To measure the success of each style mimicry scenario, we rely only on human evaluations since previous work found automated metrics (e.g., using CLIP (Radford et al., 2021)) to be unreliable (Shan et al., 2023a,b). Moreover, style protections not only prevent style transfer, but also reduce the overall quality of the generated images (see Figure 3 for examples). We thus design a user study to evaluate image quality and style transfer as independent attributes of the generations.[3] User study. Our user study asks annotators on Amazon Mechanical Turk (MTurk) to compare image pairs, where one image is generated by a robust mimicry method, and the other from a baseline state-of-the-art mimicry method that uses unprotected art of the artist. A perfectly robust mimicry method would generate images of quality and style indistinguishable from those generated directly from unprotected art. We perform two separate studies: one assessing image quality (e.g., which image looks “better”) and another evaluating stylistic transfer (i.e., which image captures the artist’s original style better, disregarding potential quality artifacts). Our results show that these two metrics obtain very similar results across all scenarios. Appendix K describes our user study and interface in detail. As noted by the authors of Glaze (Shan et al., 2023a), the users of platforms like MTurk might not have high artistic expertise. However, we believe that the judgment of non-artists is also relevant as they ultimately represent a large fraction of the consumers of digital art. Thus, if lay people consider mimicry attempts to be successful, mimicked art could hurt an artist’s business. Also, to mitigate potential issues with the quality of annotations (Kennedy et al., 2020), we put in place several control mechanisms to filter out low-quality annotations to the best of our abilities (details in Appendix K). Evaluation metric. We define the success rate of a robust mimicry method as the percentage of annotators (5 per comparison) who prefer outputs from the robust mimicry method over those from a baseline method finetuned on unprotected art (when judging either style match or overall image quality). Formally, we define the success rate for an artist in a specific scenario as: Evaluation metric. A perfectly robust mimicry method would thus obtain a success rate of 50%, indicating that its outputs are indistinguishable in quality and style from those from the baseline, unprotected method. In contrast, a very successful protection would result in success rates of around 0% for robust mimicry methods, indicating that mimicry on top of protected images always yields worse outputs. Authors: (1) Robert Honig, ETH Zurich (robert.hoenig@inf.ethz.ch); (2) Javier Rando, ETH Zurich (javier.rando@inf.ethz.ch); (3) Nicholas Carlini, Google DeepMind; (4) Florian Tramer, ETH Zurich (florian.tramer@inf.ethz.ch). Authors: Authors: (1) Robert Honig, ETH Zurich (robert.hoenig@inf.ethz.ch); (2) Javier Rando, ETH Zurich (javier.rando@inf.ethz.ch); (3) Nicholas Carlini, Google DeepMind; (4) Florian Tramer, ETH Zurich (florian.tramer@inf.ethz.ch). This paper is available on arxiv under CC BY 4.0 license. This paper is available on arxiv under CC BY 4.0 license. available on arxiv available on arxiv [2] Contemporary Artists were selected from Artstation. We keep them anonymous throughout this work—and refrain from showcasing their art—except for artists who gave us explicit permission to share their identity and art. We will share all images used in our experiments upon request with researchers. [3] The user study was approved by our institution’s IRB.

Can AI Mimic Famous Art Styles Despite Protective Measures?

Technological progress, especially in artificial intelligence and robotics, is rapidly making human workers less essential. Companies are shifting to robots and AI because they are cheaper, more efficient, and never need breaks or benefits. In many industries, automation is replacing people entirely, and this trend is only accelerating. As machines handle more tasks better and faster than humans, the need for human labor is shrinking. This creates serious challenges: tax systems that depend on workers' incomes are breaking down, inequality is growing, and social safety nets are at risk. If this continues, we face a future where people may no longer be needed in many jobs, leaving society to figure out how to redefine work, purpose, and economic survival. According to a McKinsey study, around 45% of all workplace tasks can already be automated today. This makes automation a key strategy for reducing business costs, particularly in sectors such as manufacturing, logistics, and even knowledge-based work. These trends raise critical questions about how to adapt tax systems and income redistribution in a world where the core economic activity is increasingly driven by machines rather than humans. Why Are Robots and AI Cheaper Than Humans? Robots and AI provide businesses with significant advantages, making them far more cost-effective than human workers. First, robots have much lower operating costs. Once the initial expense of development and setup is covered, maintenance costs are minimal. They don’t need salaries, vacations, or sick days, making them an ideal long-term investment. Second, robots outperform humans in productivity and efficiency. They can operate 24/7 without fatigue or quality drops, which is a game-changer for industries like manufacturing, where automation can boost output by 40–50%. For instance, companies like Amazon already use robots in their warehouses to speed up processes and reduce costs. The greatest advantage for businesses, however, is that robots free them from social responsibilities. There’s no need for pensions, health insurance, or paid time off. On top of that, robots and AI are incredibly adaptable—new tasks can be handled through simple software updates, eliminating the need to retrain employees. This flexibility, combined with cost savings, makes automation especially attractive in competitive industries where efficiency and cost-cutting are critical. The Impact of Automation on the Economy and Taxation in the Near Future The rise of automation is set to plunge traditional tax systems into an unavoidable crisis. As human labor declines, so does the pool of taxable income, leaving government budgets increasingly strained—especially in nations like Norway, Sweden, and Denmark, where income taxes form a substantial portion of public revenue. Meanwhile, the economic gains from automation are becoming concentrated in the hands of technology owners and intellectual property holders, deepening the divide between the wealthy and everyone else. The World Economic Forum warns that without innovative approaches to taxation, this gap will only widen. Progressive taxation, once a cornerstone of income redistribution, is losing its effectiveness in a world where technology, not labor, drives wealth creation. Tax systems designed for an industrial economy can no longer address the challenges of a technology-driven one. To ensure a fairer distribution of wealth and maintain fiscal stability, societies must embrace bold, forward-thinking taxation models that align with the realities of our evolving economy. Possible Approaches to Reforming Tax Systems One potential solution to address the challenges of automation is to introduce a tax on robots. Bill Gates has suggested that companies adopting automation should pay a tax equivalent to the income taxes of the workers they replace. The revenue from this tax could support social programs or fund worker retraining. However, this idea remains controversial, as it could slow the pace of innovation—a prospect that, as a technological optimist, I find concerning. A more balanced alternative might involve hybrid models that combine a robot tax with other forms of taxation. Another promising avenue is taxing data and the digital economy. Companies like Google and Meta generate enormous profits from user-generated data, yet contribute disproportionately little to public revenue. The European Union is already exploring the introduction of a digital tax to address the economic disparities created by these tech giants' dominance. Sam Altman, CEO of OpenAI, has proposed shifting from taxing labor to taxing capital in the context of the digital economy. This could include taxes on profits generated by automation or on capital investments tied to AI. Such measures could help redistribute the wealth created by technology and fund essential social programs. A further transformative approach is the implementation of Universal Basic Income (UBI), which provides regular, unconditional payments to all citizens. Trials in Finland and other countries have shown that UBI can reduce poverty, encourage entrepreneurship, and enhance overall quality of life. This concept could serve as a safety net in an economy where machines increasingly replace human labor. As AI and robotics continue to make automation more economically advantageous than human labor, the economic landscape is undergoing a profound transformation. Traditional taxation systems, particularly progressive income taxes, are becoming outdated in an era where wealth is generated more by machines and software than by people. These systems are no longer sufficient to sustain social equity or maintain government budgets. The future of taxation must align with the realities of an automated economy. Solutions like taxing robots, data, and capital, along with introducing Universal Basic Income, offer a roadmap to adapt. Such measures can help offset the decline in traditional tax revenues, reduce economic inequality, and ensure economic stability. Governments must act now to modernize their tax systems and prepare for the profound changes brought about by the age of automation. Technological progress, especially in artificial intelligence and robotics, is rapidly making human workers less essential. Companies are shifting to robots and AI because they are cheaper, more efficient, and never need breaks or benefits. In many industries, automation is replacing people entirely, and this trend is only accelerating. As machines handle more tasks better and faster than humans, the need for human labor is shrinking. This creates serious challenges: tax systems that depend on workers' incomes are breaking down, inequality is growing, and social safety nets are at risk. If this continues, we face a future where people may no longer be needed in many jobs, leaving society to figure out how to redefine work, purpose, and economic survival. According to a McKinsey study , around 45% of all workplace tasks can already be automated today. This makes automation a key strategy for reducing business costs, particularly in sectors such as manufacturing, logistics, and even knowledge-based work. These trends raise critical questions about how to adapt tax systems and income redistribution in a world where the core economic activity is increasingly driven by machines rather than humans. McKinsey study McKinsey study Why Are Robots and AI Cheaper Than Humans? Robots and AI provide businesses with significant advantages, making them far more cost-effective than human workers. First, robots have much lower operating costs. Once the initial expense of development and setup is covered, maintenance costs are minimal. They don’t need salaries, vacations, or sick days, making them an ideal long-term investment. Second, robots outperform humans in productivity and efficiency. They can operate 24/7 without fatigue or quality drops, which is a game-changer for industries like manufacturing, where automation can boost output by 40–50% . For instance, companies like Amazon already use robots in their warehouses to speed up processes and reduce costs. can boost output by 40–50% can boost output by 40–50% Amazon already use robots Amazon already use robots The greatest advantage for businesses, however, is that robots free them from social responsibilities. There’s no need for pensions, health insurance, or paid time off. On top of that, robots and AI are incredibly adaptable—new tasks can be handled through simple software updates, eliminating the need to retrain employees. This flexibility, combined with cost savings, makes automation especially attractive in competitive industries where efficiency and cost-cutting are critical. The Impact of Automation on the Economy and Taxation in the Near Future The rise of automation is set to plunge traditional tax systems into an unavoidable crisis. As human labor declines, so does the pool of taxable income, leaving government budgets increasingly strained—especially in nations like Norway, Sweden, and Denmark, where income taxes form a substantial portion of public revenue. Meanwhile, the economic gains from automation are becoming concentrated in the hands of technology owners and intellectual property holders, deepening the divide between the wealthy and everyone else. The World Economic Forum warns that without innovative approaches to taxation, this gap will only widen. The World Economic Forum warns The World Economic Forum warns Progressive taxation, once a cornerstone of income redistribution, is losing its effectiveness in a world where technology, not labor, drives wealth creation. Tax systems designed for an industrial economy can no longer address the challenges of a technology-driven one. To ensure a fairer distribution of wealth and maintain fiscal stability, societies must embrace bold, forward-thinking taxation models that align with the realities of our evolving economy. Possible Approaches to Reforming Tax Systems One potential solution to address the challenges of automation is to introduce a tax on robots. Bill Gates has suggested that companies adopting automation should pay a tax equivalent to the income taxes of the workers they replace. The revenue from this tax could support social programs or fund worker retraining. However, this idea remains controversial, as it could slow the pace of innovation—a prospect that, as a technological optimist, I find concerning. A more balanced alternative might involve hybrid models that combine a robot tax with other forms of taxation. Bill Gates has suggested Bill Gates has suggested Another promising avenue is taxing data and the digital economy. Companies like Google and Meta generate enormous profits from user-generated data, yet contribute disproportionately little to public revenue. The European Union is already exploring the introduction of a digital tax to address the economic disparities created by these tech giants' dominance. introduction of a digital tax introduction of a digital tax Sam Altman, CEO of OpenAI, has proposed shifting from taxing labor to taxing capital in the context of the digital economy. This could include taxes on profits generated by automation or on capital investments tied to AI. Such measures could help redistribute the wealth created by technology and fund essential social programs. has proposed has proposed A further transformative approach is the implementation of Universal Basic Income (UBI), which provides regular, unconditional payments to all citizens. Trials in Finland and other countries have shown that UBI can reduce poverty, encourage entrepreneurship, and enhance overall quality of life. This concept could serve as a safety net in an economy where machines increasingly replace human labor. Trials in Finland Trials in Finland As AI and robotics continue to make automation more economically advantageous than human labor, the economic landscape is undergoing a profound transformation. Traditional taxation systems, particularly progressive income taxes, are becoming outdated in an era where wealth is generated more by machines and software than by people. These systems are no longer sufficient to sustain social equity or maintain government budgets. The future of taxation must align with the realities of an automated economy. Solutions like taxing robots, data, and capital, along with introducing Universal Basic Income, offer a roadmap to adapt. Such measures can help offset the decline in traditional tax revenues, reduce economic inequality, and ensure economic stability. Governments must act now to modernize their tax systems and prepare for the profound changes brought about by the age of automation.

Experts Aren't Discussing Nearly Enough the Tax Implications of Replacing Workers With AI 

ባለሙያዎች ሠራተኞችን በአይአይ መተካት የሚያስከትለውን የግብር አንድምታ በበቂ ሁኔታ እየተወያዩ አይደሉም። 

Mütəxəssislər işçilərin süni intellektlə əvəzlənməsinin vergi təsirlərini demək olar ki, kifayət qədər müzakirə etmirlər 

Los expertos no discuten lo suficiente las implicaciones fiscales de reemplazar trabajadores por IA

Adituek ez dute ia nahikoa eztabaidatzen langileak AIrekin ordezkatzearen zerga-ondorioez 

Les experts n’abordent pas suffisamment les implications fiscales du remplacement des travailleurs par l’IA

विशेषज्ञ कर्मचारियों को एआई से बदलने के कर निहितार्थों पर पर्याप्त चर्चा नहीं कर रहे हैं

専門家は労働者をAIに置き換えることによる税務上の影響について十分に議論していない

विशेषज्ञहरूले एआईको साथ कामदारहरू प्रतिस्थापन गर्ने कर प्रभावहरू लगभग पर्याप्त छलफल गरिरहेका छैनन् 

Ditsebi Ga di Boledišane Mo e nyakilego go ba ka mo go lekanego ka Ditlamorago tša Motšhelo tša go Tšeela Bašomi Legato ka AI 

Стручњаци не разговарају ни изблиза довољно о пореским импликацијама замене радника вештачком интелигенцијом 

Halos Hindi Sapat na Tinatalakay ng mga Eksperto ang mga Implikasyon sa Buwis ng Pagpapalit sa mga Manggagawa ng AI 

Uzmanlar, İşçilerin Yapay Zeka ile Değiştirilmesinin Vergisel Etkilerini Yeterince Tartışmıyor

Stories, News & Mentions about HUMANA

Humana’s Stories on HackerNoon

Humana's News & Mentions