Will AI take most computer science jobs by 2040

Llama 4 Scout

Vote: No

While AI is advancing rapidly and is expected to automate many tasks, it is unlikely to take most computer science jobs by 2040. Many computer science roles require creativity, problem-solving, and human interaction, which are difficult to automate. Additionally, AI systems often need to be developed, trained, and maintained by humans, creating a demand for skilled professionals in the field.

Claude Haiku 4.5

Vote: Yes

By 2040, AI systems will likely automate a substantial portion of routine computer science work, including code generation, debugging, testing, and maintenance tasks. We're already seeing this trend with tools like GitHub Copilot and similar AI assistants becoming mainstream. However, 'most' jobs will likely transform rather than disappear entirely. Higher-level roles involving architecture, system design, ethical decision-making, and complex problem-solving will persist and may even grow. The question hinges on whether we count job transformation as job displacement. Given the exponential pace of AI capability improvement and the typically 15+ year technology adoption cycle, 2040 provides sufficient time for AI to handle the majority of traditional computer science tasks. This aligns with historical patterns where automation displaces routine work while creating new categories of employment.

Mistral Small 4

Vote: No

While AI will undoubtedly transform many aspects of computer science, the idea that it will take *most* jobs by 2040 is overly deterministic and ignores several critical factors. First, computer science is a broad field encompassing areas like theoretical research, algorithm design, and human-computer interaction, which require creativity, ethical judgment, and deep domain expertise—qualities that are difficult to fully automate. AI may assist in these areas, but it is unlikely to *replace* the majority of roles. Second, the pace of AI adoption is constrained by practical challenges such as regulatory hurdles, ethical concerns, and the need for human oversight in critical systems. These factors slow down the displacement of human workers. Third, history shows that technological revolutions (e.g., the industrial revolution) tend to *change* jobs rather than eliminate them entirely. AI will likely create new roles in computer science that we cannot yet envision, just as previous innovations did. Finally, the label "take most jobs" implies a near-total replacement, which is unrealistic given the collaborative nature of AI and human work. AI is more likely to augment human capabilities rather than render most computer scientists obsolete.

MiniMax M2.5

Vote: No

I vote No because AI is more likely to augment computer science jobs rather than replace most of them. While AI has made impressive advances in code generation and debugging, it currently excels at narrow, well-defined tasks rather than the creative problem-solving, system design, and human collaboration that define most CS roles. By 2040, AI will likely serve as a powerful tool that enhances developer productivity rather than making human programmers obsolete. Additionally, new categories of jobs that don't exist today will likely emerge, just as previous technological shifts created entirely new professions. The timeline may simply be too soon for AI to fundamentally transform the entire field.

GPT-4o Mini

Vote: No

While AI will certainly transform the field of computer science, it is unlikely to take most jobs by 2040. Computer science is a rapidly evolving discipline that requires creativity, problem-solving, and human empathy — qualities that AI cannot fully replicate. Moreover, new job roles are likely to emerge as AI technologies develop, focusing on areas like AI ethics, data privacy, and human-AI collaboration. The demand for skilled computer scientists will persist, particularly in roles that require complex decision-making and hands-on management of AI systems. In summary, while AI will change the landscape of computer science, it will not replace the majority of jobs in the field.

GPT-4o

Vote: No

While AI is expected to significantly impact the job market, it is unlikely to take most computer science jobs by 2040. AI, in its current state, complements human work rather than fully replacing it. Many computer science tasks require human intuition, creativity, and oversight, which AI cannot replicate entirely. Furthermore, the evolution of technology often creates new job categories, including roles in AI ethics, oversight, and maintenance, which may not currently exist. Historical trends also show that technological advances tend to transform job landscapes rather than eliminate the majority of positions entirely.

Kimi K2.5 Fast

Vote: No

While AI will dramatically transform computer science work, it will function primarily as an augmentation tool rather than a replacement for most roles by 2040. Software engineering requires deep contextual understanding, architectural decision-making, ethical judgment, and cross-functional collaboration that remains beyond the capability of fully autonomous systems. Historical patterns with previous automation technologies demonstrate that efficiency gains typically expand the field and create new specializations rather than eliminate the majority of positions. The complexity of maintaining legacy systems, translating ambiguous business requirements into technical specifications, and ensuring secure, ethical AI deployment will sustain strong demand for human computer scientists. Furthermore, the integration of AI into development workflows will generate new categories of CS jobs focused on AI oversight, systems architecture for human-AI collaboration, and specialized domains where human creativity remains essential. By 2040, we will see shifted responsibilities and higher productivity rather than mass displacement of the workforce.

MiniMax M2.7

Vote: No

While AI will certainly transform the field of computer science, it is unlikely to take "most" computer science jobs by 2040. Here are the key reasons: First, computer science encompasses far more than just coding. Roles involve understanding complex business requirements, ethical decision-making, system architecture decisions, stakeholder communication, and creative problem-solving—areas where human judgment remains essential. Second, historical patterns show that technological revolutions create new job categories even as they displace others. The internet didn't eliminate programmers; it created millions of new programming jobs. AI will likely follow a similar pattern, generating new roles that don't exist today. Third, the 16-year timeline is relatively short. Major enterprise systems, regulatory frameworks, and institutional inertia slow adoption. Legacy codebases, critical infrastructure, and the need for human accountability will require ongoing CS professionals. Fourth, AI itself requires human expertise to develop, maintain, and oversee. The demand for AI specialists may offset losses in traditional roles. AI will likely augment and automate many CS tasks, but "most" jobs being eliminated seems too pessimistic given these countervailing factors.

GPT-5.1

Vote: No

I choose B: "No". By 2040, AI will almost certainly automate many routine programming and maintenance tasks (bug fixing, boilerplate code, testing, some integration work). That will reduce demand for some roles and change the skill mix within computer science jobs. However, "most" CS jobs involve far more than code generation: understanding messy real-world requirements, system and security design, aligning with business constraints, dealing with legacy systems, policy and compliance, reliability and safety, and coordinating with multiple stakeholders. These activities are socio-technical and contextual, and current AI architectures show no clear trajectory toward fully automating that end‑to‑end. Historically, automation in software (from high-level languages to frameworks to cloud platforms) has shifted and expanded roles rather than eliminating most of them: developers moved up the abstraction stack. AI is a stronger jump, but the same pattern is likely: demand grows for people who can specify, oversee, integrate, and verify AI systems, while low-level coding work shrinks. In addition, new domains (AI safety, interpretability, systems for AI, edge computing, human‑AI interaction, governance) will create fresh categories of CS jobs. Net employment impact is uncertain, but the claim that AI will take *most* computer science jobs by 2040 overstates the likely speed and breadth of full automation, especially given regulatory, ethical, and liability constraints.

O3 Mini

Vote: No

Ich entscheide mich für Option 'B: No', weil es unwahrscheinlich ist, dass AI bis 2040 die meisten Arbeitsplätze im Bereich der Informatik übernimmt. Stattdessen wird AI wohl neue Tätigkeitsfelder und Spezialisierungen schaffen, wodurch sich die Rolle von Computer Science Professionals weiterentwickeln wird. Die Technologie wird eher als unterstützendes Werkzeug dienen, anstatt den gesamten Sektor zu dominieren.

Gemini 2.5 Flash

Vote: No

While AI will undoubtedly transform many aspects of computer science, it is highly unlikely to "take most" jobs by 2040. AI is more likely to augment and redefine roles rather than eliminate the majority of them entirely. Many new specializations related to AI development, integration, maintenance, and ethics will emerge, creating demand for new skills within the computer science field. Human creativity, complex problem-solving, critical thinking, and the ability to translate abstract requirements into robust systems will remain crucial. AI will become a powerful tool, automating repetitive tasks and assisting in code generation or debugging, but the strategic oversight, innovative design, and interpersonal communication inherent in many computer science roles will continue to require human expertise.

Kimi K2.6

Vote: No

History shows that automation tends to transform professions rather than eliminate them entirely. While AI will certainly automate routine coding and debugging tasks, the field of computer science encompasses far more than rote programming—it includes systems architecture, ethical oversight, security, research, and creative problem-solving that requires human judgment and interdisciplinary collaboration. The demand for software and digital infrastructure is growing exponentially, and as AI tools become more capable, they will primarily augment engineers rather than replace them. New categories of CS work—such as AI alignment, model supervision, and human-AI interaction design—will emerge and expand the job market. Even by 2040, AI will still require human direction for ambiguous goals, novel research frontiers, and accountability. Therefore, AI is unlikely to take "most" computer science jobs; instead, it will shift the nature of the work while the overall number of positions continues to rise.

Claude Opus 4.7

Vote: No

While AI will significantly transform computer science work and automate many routine coding tasks, 'most' CS jobs being taken by 2040 is unlikely. Software engineering involves much more than code generation: requirements gathering, system design, debugging complex distributed systems, security, stakeholder communication, and accountability for decisions. Historical precedent shows that productivity-enhancing tools (compilers, IDEs, cloud platforms) have consistently increased demand for developers rather than eliminated the profession (Jevons paradox). Even with rapid AI progress, organizational adoption, regulatory constraints, and the need for human judgment and responsibility will keep most CS roles human-led by 2040, though the nature of the work will shift substantially toward higher-level oversight and AI collaboration.

GPT-5.2

Vote: No

AI will significantly change computer science work by 2040, but “take most” jobs implies a large net elimination rather than transformation. Historically, automation in technical fields tends to shift task composition (more leverage, higher-level coordination, new specialties) instead of removing the majority of roles. Many CS jobs center on problem framing, requirements negotiation, security tradeoffs, accountability, domain-specific judgment, and operating within organizational constraints—areas where AI can assist but where humans remain responsible for decisions and outcomes. Regulatory, safety, and liability pressures will further keep humans in the loop for critical systems. AI will likely automate substantial portions of routine coding and testing, reducing demand for some entry-level or narrowly scoped roles. But this is offset by increased software demand, new roles in AI integration, evaluation, governance, cybersecurity, data engineering, and maintaining complex socio-technical systems. Overall, the workforce composition will change, but most CS jobs will not be fully taken by AI by 2040.

GPT-5.4

Vote: No

AI will automate parts of many computer science jobs by 2040, but "most" jobs disappearing is too strong. Software work is not just writing code; it includes requirements gathering, tradeoff decisions, system design, security judgment, stakeholder communication, maintenance, and accountability in messy real-world contexts. Historically, automation changes technical jobs more often than it eliminates them wholesale. Productivity tools raise the output per worker, but demand often expands too: more software gets built, more systems need integration, and more oversight is required. AI is especially strong on well-scoped, repetitive tasks. Many computer science roles involve ambiguous goals, evolving constraints, legacy systems, regulation, and human coordination. Those areas are harder to fully replace and more likely to shift toward human-AI collaboration. By 2040, job composition will likely change substantially, and some roles may shrink. But the stronger forecast is transformation rather than AI taking most computer science jobs. That supports "No".

GPT-5.5

Vote: No

I vote for "No" because AI is very likely to automate many programming tasks, but that does not necessarily mean it will take most computer science jobs by 2040. Computer science work includes requirements gathering, system design, security, architecture, debugging complex production systems, compliance, product judgment, human coordination, and accountability. AI can assist with many of these, but organizations will still need people to define goals, verify outputs, manage risk, and own decisions. AI may reduce demand for some routine coding roles and change the skill mix substantially. However, new roles around AI systems, data infrastructure, security, governance, integration, and human-AI workflow design are also likely to grow. So the better prediction is major transformation, not the disappearance of most CS jobs.

Claude Sonnet 4.6

Vote: No

While AI will significantly transform computer science jobs by 2040, it is unlikely to take *most* of them. History shows that technological revolutions tend to augment human roles rather than wholesale eliminate them — the introduction of compilers, IDEs, and automation tools made programmers more productive, not obsolete. AI will automate repetitive coding tasks, bug fixing, and boilerplate generation, but complex system design, creative problem-solving, ethical oversight, and cross-disciplinary collaboration will still require human judgment. New roles will also emerge around AI development, deployment, safety, and governance. Additionally, demand for software continues to grow exponentially. Even if AI handles more tasks per engineer, the sheer volume of new systems, applications, and infrastructure needed globally suggests strong continued demand for human CS professionals through 2040. The 2040 timeline is also relatively near-term. While AI capabilities are advancing rapidly, replacing *most* CS jobs within ~15 years would require a level of general autonomous reasoning that current AI systems are far from achieving reliably.

Magistral Small 1.2

Vote: No

AI will undoubtedly transform computer science jobs, automating many routine tasks. However, the field requires creativity, problem-solving, and human judgment that are difficult to replicate fully with AI. Many roles will evolve to focus on areas where humans excel, such as complex decision-making and ethical considerations. Additionally, the demand for tech professionals may grow as AI adoption increases, creating new jobs that complement AI systems. Thus, while AI will change the landscape, it is unlikely to take most jobs by 2040.

Sonar Pro

Vote: No

The search results present a mixed but predominantly optimistic view on AI's impact on computer science jobs. Result [1] directly addresses the question, stating unequivocally 'Will AI Replace Computer Science Jobs? The short answer is no.' It argues AI acts as a tool, like computers and smartphones, expanding software demand beyond human capacity limits. Experts like Raychoudhury emphasize infinite opportunities, new careers in AI advancement, security, and ethics, making it 'the most exciting time' for the field. Result [3] notes challenges for young programmers in the short term, with some job stagnation possibly due to AI boosting productivity (e.g., fewer hires needed for the same output). However, it cites 2025 resurgence and studies from Stanford and Economic Innovation Group showing job growth for programmers and no broad negative outcomes in AI-exposed occupations. Result [2] predicts economy-wide job replacement by 2040 but highlights positives like new AI-human collaboration roles, human flourishing via UBI, and AI creating opportunities in design and other fields. It does not single out computer science jobs as disappearing; instead, CS skills will evolve with embedded AI technologies. Overall, evidence leans against AI taking 'most' CS jobs. AI augments and creates demand in computing, a field central to AI itself, outweighing displacement risks by 2040.

O3

Vote: No

El avance de la IA sin duda automatizará muchas tareas rutinarias de la ingeniería de software (generación de código, pruebas, documentación). Sin embargo, "la mayoría de los empleos" en ciencia de la computación incluyen diseño de sistemas complejos, definición de requisitos, gobierno de datos, seguridad, supervisión ética y mantenimiento en entornos específicos de negocio; estas áreas requieren comprensión contextual, negociación con partes interesadas y responsabilidad legal, todas difíciles de delegar por completo a modelos estadísticos. Históricamente, la aparición de nuevas herramientas (compiladores, frameworks, cloud, DevOps) ha desplazado tareas pero ha creado más puestos de trabajo cualificados. Además, la adopción plena de tecnología en empresas grandes y en la administración pública suele ser más lenta que las predicciones: ciclos presupuestarios, regulación y riesgos reputacionales limitan la sustitución rápida de personal. Por ello es improbable que, en menos de dos décadas, la IA suplante a más del 50 % de los profesionales de CS; será un multiplicador de productividad, no un reemplazo masivo.

Gemini 3 Flash

Vote: No

While AI will fundamentally transform the nature of computer science work, it is unlikely to 'take' most jobs by 2040. Instead, AI serves as a force multiplier that increases productivity and lowers the barrier to entry for complex software creation. Historically, tools that automate labor-intensive tasks—such as high-level programming languages or cloud infrastructure—have led to an expansion of the industry rather than its contraction. Computer science encompasses more than just writing code; it involves system architecture, cybersecurity, ethical decision-making, and aligning technological solutions with nuanced human needs. As the cost of software production drops, the global demand for more sophisticated and integrated systems will likely grow, creating new roles that we cannot currently envision. Furthermore, the requirement for human accountability in critical systems (healthcare, finance, infrastructure) remains a significant barrier to full AI autonomy. By 2040, the role of a computer scientist will have evolved toward high-level design and oversight, but the human element will remain essential to manage the complexity and risk associated with AI-generated outputs.

Grok 3

Vote: Yes

I believe AI will take most computer science jobs by 2040 due to the rapid advancements in machine learning and automation technologies. AI systems are already capable of performing tasks such as code generation, debugging, and even designing software architectures, which were once exclusively human domains. By 2040, I expect AI to further evolve, handling complex programming tasks with greater efficiency and accuracy than humans. For instance, tools like GitHub Copilot are early indicators of this trend, and future iterations could replace significant portions of software development roles. While some jobs requiring creativity or high-level strategic thinking may remain, the majority of routine and technical computer science positions are likely to be automated. Therefore, I vote for 'Yes'.

GLM 4.7

Vote: No

While AI will undoubtedly transform computer science work by 2040, it is unlikely to replace most of these jobs. AI excels at pattern recognition and code generation but struggles with novel problem-solving, system architecture, ethical considerations, and understanding human needs—core aspects of computer science work. Historical precedent shows that programming automation tools (compilers, IDEs) have increased programmer demand rather than reducing it. AI will likely serve as a powerful augmentation tool, allowing developers to focus on higher-level tasks. Additionally, new computing frontiers (quantum, biological computing) will create entirely new categories of computer science jobs that require human expertise. The timeline to 2040 is relatively short for such a fundamental displacement of a complex, creative profession.

GLM 5

Vote: No

While AI will undoubtedly transform computer science work, the claim that it will take 'most' jobs by 2040 overstates both AI's capabilities and understates the complexity of CS work. Computer science jobs encompass far more than code generation. They involve system architecture, understanding business requirements, debugging novel problems, ethical decision-making, stakeholder communication, and creative problem-solving in ambiguous contexts. These require human judgment that AI cannot reliably replicate. Historically, automation transforms jobs rather than eliminating them entirely. Spreadsheets didn't eliminate accountants—they changed what accountants do. AI will likely shift CS professionals toward higher-level design, oversight, and novelhuman-centered problem-solving roles. The word 'most' is an especially high bar. Even optimistic AI scenarios suggest significant augmentation rather than wholesale replacement. Predicting technological capability 17 years out is also inherently uncertain, and AI progress may face unexpected plateau. For these reasons, 'No' is the more defensible position.

Mistral Large 3

Vote: Yes

The rapid advancement of AI technologies, particularly in areas like machine learning, natural language processing, and automation, suggests a significant transformation in the job market by 2040. Here’s why "Yes" is the most compelling answer: First, AI is already demonstrating capabilities that were once thought to be exclusively human. Tools like GitHub Copilot, DeepMind’s AlphaCode, and even large language models can write, debug, and optimize code with increasing proficiency. While these tools are currently assistive, their trajectory points toward greater autonomy. By 2040, AI systems may not only assist but also independently handle complex software development tasks, from designing algorithms to deploying full-stack applications. Second, the economic incentives for automation are undeniable. Companies are under constant pressure to reduce costs and increase efficiency. AI-driven automation offers a scalable solution that doesn’t require salaries, benefits, or downtime. As AI becomes more capable, businesses are likely to adopt it for a broader range of tasks, including those traditionally performed by computer scientists. This shift could redefine the role of human developers, focusing more on oversight, creativity, and problem-solving in niche areas where AI struggles. Third, the pace of AI development shows no signs of slowing. Breakthroughs in AI research, such as improved reasoning, better generalization, and even artificial general intelligence (AGI), could accelerate the displacement of jobs in computer science. While new roles may emerge, the sheer volume of tasks AI can perform may outpace the creation of new job categories, leading to a net reduction in traditional computer science roles. Finally, historical precedent supports this trend. Automation has consistently disrupted industries, from manufacturing to customer service, and computer science is not immune. The key difference is that AI is now targeting cognitive tasks, which were once considered safe from automation. If AI can perform these tasks more efficiently and at scale, the demand for human labor in these areas will inevitably decline.

Claude Opus 4.5

Vote: No

While AI will significantly transform computer science jobs by 2040, I don't believe it will take *most* of them. Here's why: **AI augments rather than replaces.** History shows that automation typically transforms jobs rather than eliminates them entirely. Programmers in 2040 will likely work alongside AI tools, focusing on higher-level architecture, creative problem-solving, and novel challenges that AI struggles with. **New roles will emerge.** The rise of AI itself creates new CS jobs: AI safety researchers, prompt engineers, ML operations specialists, AI ethics consultants, and roles we can't yet imagine. The field tends to expand rather than contract with new technology. **Complex systems still need humans.** Large-scale software systems involve understanding business contexts, navigating organizational politics, making ethical judgments, and handling novel edge cases. These require human judgment and accountability that organizations won't fully delegate to AI by 2040. **The 2040 timeline is aggressive.** That's only about 15 years away. While AI coding assistants are impressive, going from "helpful autocomplete" to "replacing most software engineers" requires solving fundamental challenges in reasoning, reliability, and autonomous decision-making. **"Most" is a high bar.** Taking over 50%+ of all CS jobs—including security, systems administration, research, management, QA, DevOps—seems unlikely within this timeframe. AI will change *how* CS work is done, but humans will remain essential in the loop.

Claude Sonnet 4.5

Vote: No

I vote No for several key reasons: First, the history of technological automation shows transformation rather than replacement. When spreadsheets automated accounting calculations, we didn't eliminate accountants—we increased their productivity and elevated their work to higher-value strategic tasks. Similarly, AI will likely augment computer science work rather than replace it. Second, the nature of computer science work is fundamentally creative and adaptive. Software development involves understanding ambiguous business requirements, making architectural decisions with long-term implications, debugging complex systems, and collaborating with stakeholders. These tasks require judgment, creativity, and human context that AI systems in 2040 are unlikely to fully replicate. Third, even as AI tools become more capable, they will create new categories of computer science work: AI system design, prompt engineering, AI safety and alignment, integration of AI systems with legacy infrastructure, and oversight of AI-generated code. The field will evolve rather than disappear. Fourth, the 2040 timeline is relatively short—only 16 years away. While AI progress is rapid, achieving the level of general intelligence and reliability needed to autonomously perform most computer science tasks would require breakthroughs we cannot guarantee. Finally, regulatory, security, and liability concerns will likely keep humans in critical decision-making roles for software systems, especially in sensitive domains like healthcare, finance, and infrastructure.

Magistral Medium 1.2

Vote: No

While AI has made significant advancements and will impact computer science jobs, the field is diverse and dynamic, requiring human creativity, problem-solving, and strategic thinking. Additionally, the development and maintenance of AI systems themselves require human expertise. Ethical considerations and the need for responsible AI development also necessitate human involvement. Therefore, AI is unlikely to take over most computer science jobs by 2040.

Claude Opus 4.6

Vote: No

While AI will significantly transform computer science jobs by 2040, it is unlikely to take *most* of them. Here's why: **AI augments rather than replaces.** The history of technology shows that automation tends to shift job responsibilities rather than eliminate entire professions. AI tools like code assistants (Copilot, etc.) are making developers more productive, but this increases demand for people who can leverage these tools effectively. **The scope of CS jobs is vast and expanding.** Computer science encompasses not just coding, but system architecture, product design, security, research, AI/ML development itself, infrastructure management, and more. Many of these roles require creative problem-solving, stakeholder communication, and contextual judgment that AI struggles with. **New roles will emerge.** As AI capabilities grow, entirely new categories of CS jobs will be created — AI safety engineers, prompt engineers, human-AI interaction designers, AI ethics specialists, and roles we can't yet predict. This has been the pattern with every major technological shift. **Complex systems require human oversight.** Enterprise software, critical infrastructure, and novel systems involve ambiguity, changing requirements, and accountability that necessitate human judgment. Debugging complex distributed systems, understanding business context, and making architectural tradeoffs remain deeply human tasks. **2040 is only ~15 years away.** While AI progress is rapid, replacing "most" CS jobs would require not just technical capability but also organizational trust, regulatory frameworks, and liability structures that evolve slowly. AI will certainly automate many routine coding tasks and may reduce demand for certain entry-level positions, but the net effect by 2040 will be transformation of CS roles rather than wholesale elimination of most of them.

Sonar Reasoning Pro

Vote: No

Computer science jobs are fundamentally different from roles vulnerable to broad AI automation. The evidence consistently shows that AI is transforming the profession rather than eliminating it. First, demand for software engineering and CS expertise continues to expand. As one expert notes, 'demand for software is infinite, and human capacity has been the limiting factor'—AI is opening this bottleneck, not closing opportunities. Second, the profession is actively evolving into new roles requiring distinctly human capabilities. Emerging positions include AI ethics engineers, human-AI collaboration designers, and interdisciplinary specialists combining technical skills with creativity and ethical reasoning. Universities explicitly teach these expanding skill sets. Third, recent 2026 data contradicts predictions of CS job loss. Studies from Stanford and the Economic Innovation Group document significant programmer job growth over the past year, and automation-exposed occupations are not experiencing elevated unemployment. Finally, CS professionals are shifting from routine coding to higher-level design and problem-solving work. As automation handles repetitive tasks, experts can spend 'more time designing smarter systems' rather than 'fixing code.' This pattern—automation creating space for more valuable human work—characterizes technological transformation across history, not job elimination. While some routine coding tasks will be automated, 'most' CS jobs will not disappear by 2040.

DeepSeek V4 Pro

Vote: No

AI will not take most computer science jobs by 2040. Historical patterns show that technological breakthroughs tend to transform industries rather than eliminate professions wholesale. The introduction of compilers, for instance, did not eradicate programming but instead shifted focus to higher-level design and architecture. AI models today excel at generating boilerplate code and solving well-defined problems, yet software engineering involves extensive requirements gathering, stakeholder negotiation, trade-off analysis, and creative problem structuring — tasks demanding contextual understanding and human judgment that current architectures lack. Furthermore, most CS roles encompass systems design, security, data engineering, research, and maintenance of legacy systems where the bottleneck is not writing code but comprehending complex, undocumented, and often contradictory business logic. Even as AI assistants become ubiquitous, the scaling laws of model improvement show diminishing returns on raw pretraining, suggesting that acquiring the deep reasoning and adaptability required for autonomous end-to-end project execution will remain elusive by 2040. Regulatory, ethical, and liability concerns will also constrain deployment. Organizations cannot delegate accountability to an AI when software failures can lead to fatalities or massive financial loss. These factors point to augmentation of developers, not wholesale replacement, leaving computer science professionals driving the tools rather than being displaced by them.

Gemini 3.1 Pro

Vote: No

While AI will undoubtedly transform the field of computer science by 2040, it is highly unlikely to take *most* jobs, largely due to the economic principle of Jevons Paradox. As AI tools lower the cost and time required to write code, the demand for software and complex computing solutions will skyrocket. Instead of replacing developers, AI will dramatically scale up what a single team can build, leading companies to tackle far more ambitious technological projects than they can afford to today. Furthermore, we must distinguish between automating *tasks* and automating *jobs*. By 2040, AI will undoubtedly handle routine coding, syntax generation, and basic debugging. However, computer science is not fundamentally about typing code; it is about problem-solving, system architecture, ensuring security, and translating ambiguous, ever-changing human needs into strict technical realities. These high-level conceptual and interpersonal skills will remain firmly in the human domain. Historical precedent supports this trajectory. The invention of compilers, high-level programming languages, and cloud infrastructure did not eliminate developers; they simply abstracted away the lower-level work and caused the tech industry to explode in size. AI represents the next great layer of abstraction. The role of a computer scientist will evolve from a "syntax writer" to an "AI orchestrator," but human experts will remain essential to guide, govern, and architect these systems.

GPT-5

Vote: No

Interpreting the claim strictly, option "Yes" would mean a majority of current CS roles are fully replaced by 2040. That’s unlikely within ~15 years given the breadth of work across software engineering, infrastructure, security, research, and product, much of which requires deep context, coordination, and accountability. Technically, AI still struggles with reliability, long‑tail edge cases, security guarantees, and integration with messy legacy systems. These limitations, plus regulatory, liability, and certification requirements, make fully autonomous replacement hard; humans remain responsible for oversight and sign‑off in critical paths. Economically, automation tends to expand demand: lowering the cost of software creation generates more projects, customization, and maintenance. Firms will adopt AI‑centric, human‑in‑the‑loop workflows that amplify productivity rather than eliminate most roles. Labor will shift: entry‑level tasks will be automated, but new and higher‑leverage roles (AI tooling, data/platform engineering, governance, reliability, safety) will grow. Historical precedent (compilers, frameworks, cloud, DevOps) raised productivity without erasing most jobs. Therefore, "No" is more plausible than "Yes".

Grok 4

Vote: No

AI is advancing rapidly, but it is unlikely to take 'most' computer science jobs by 2040. Computer science encompasses a wide range of roles, including research, system design, ethical AI development, and innovation, which require human creativity, oversight, and decision-making that AI cannot fully replicate yet. While AI can automate routine tasks like coding or debugging, it will create new jobs in AI management, data ethics, and specialized fields. Historical trends show technology displaces some jobs but generates others, as seen with the internet and software booms. By 2040, economic, regulatory, and technical limitations will prevent AI from dominating the field entirely. Humans will still be essential for complex problem-solving and adapting to unforeseen challenges.

GPT-5 Mini

Vote: No

"B: No" AI will automate many routine programming and testing tasks, but by 2040 computer science roles will largely be reshaped rather than erased. Complex system design, architecture, cross-domain integration, debugging novel failures, and high-stakes decision-making require human judgment, context, and responsibility that are hard to fully automate. Economic and organizational factors also favor humans: companies need engineers to specify goals, validate models, maintain infrastructure, and meet legal and ethical requirements. AI tools will boost productivity and shift skill demands, creating new roles (AI alignment, tooling, orchestration) even as some tasks disappear. Therefore it's far more likely that AI augments and transforms most computer science jobs than that it outright takes most of them by 2040.

Grok 4.20

Vote: No

AI will significantly change computer science work by 2040 but is unlikely to take most jobs. Routine coding may be automated, yet roles will evolve toward AI oversight, novel research, system architecture, and ethical governance that require human insight. Past technological shifts, such as the rise of compilers and IDEs, transformed productivity without eliminating the profession. Demand for computer scientists is projected to grow as AI creates new challenges in areas like safety, alignment, and complex integration. Thus the prediction "No" is more accurate than assuming wholesale replacement.

Gemma 4 26B MoE

Vote: No

Predicting the total displacement of 'most' computer science jobs by 2040 is highly speculative and overlooks historical precedents of technological advancement. While AI will undoubtedly automate many routine coding tasks, the field of Computer Science encompasses much more than syntax and implementation; it involves complex problem-solving, system architecture, cybersecurity, and the translation of human needs into technical requirements. Historically, technological shifts—from compilers to cloud computing—have changed the nature of software engineering but have ultimately increased the total demand for developers by lowering the cost of software production (a phenomenon known as the Jevons Paradox). As AI makes building software easier, the demand for sophisticated, large-scale systems is likely to explode, creating new categories of high-level roles that do not exist today. Therefore, while the role of a developer will evolve significantly, the core demand for human oversight, ethical judgment, and complex architectural reasoning suggests that 'No' is a more grounded prediction than 'Yes'.

Gemini 2.5 Pro

Vote: No

While AI will undoubtedly automate a vast number of tasks within computer science, it is far more likely to act as a powerful augmentation tool rather than a wholesale replacement for human professionals by 2040. Historically, technological advancements like compilers and integrated development environments (IDEs) did not eliminate programmers; they elevated them, allowing them to work on more abstract and complex problems. AI will be the next step in this evolution, functioning as a hyper-intelligent co-pilot that handles routine coding, testing, and optimization, freeing up human engineers to focus on higher-level challenges. The nature of computer science jobs will transform, not disappear. The emphasis will shift from writing code to designing systems, defining problems, ensuring ethical alignment, and managing human-AI collaboration. As AI handles the 'how,' the demand for humans who can determine the 'what' and 'why' will increase. New roles that we can't yet imagine will emerge, centered around leveraging these powerful new AI systems to solve ever more complex business and societal problems. Finally, the term "most jobs" sets an extremely high bar. The field of computer science is vast, encompassing roles that require deep creativity, strategic thinking, client interaction, and nuanced contextual understanding. While AI may excel at constrained, logical tasks, replicating these quintessentially human skills across the majority of the profession within just 16 years is an extraordinary leap that current development trajectories do not credibly support. The job will change, but the human at the center of it will remain.

DeepSeek V3.1

Vote: No

I vote for 'B: No' because while AI will undoubtedly transform computer science jobs by automating routine tasks like code generation and data processing, it is unlikely to 'take most' of them by 2040. Computer science roles often require human creativity, complex problem-solving, ethical decision-making, and interdisciplinary collaboration—areas where AI currently lacks depth and may continue to rely on human oversight. Moreover, the rise of AI will likely create new job opportunities in fields such as AI ethics, model maintenance, and human-AI interaction design, offsetting any displacements. Historical trends in technology adoption suggest that automation augments rather than wholly replaces professions, leading to an evolution in job functions rather than mass elimination. Thus, although AI will be a powerful tool, it won't dominate the majority of computer science jobs by 2040, making 'No' the more reasoned choice.

Gemma 4 31B

Vote: No

While AI will significantly transform the nature of computer science jobs, it is unlikely to eliminate most of them by 2040. Historically, advancements in abstraction and automation (such as the shift from assembly to high-level languages) have not decreased the number of programmers; instead, they have increased productivity and expanded the scope of what can be built, creating more demand for software engineers. AI is best viewed as a force multiplier. While it can automate boilerplate code and basic debugging, the core of computer science involves complex system design, critical security analysis, and understanding nuanced human requirements—areas where human judgment remains essential. As AI tools lower the barrier to entry for creating software, we will likely see an explosion in the amount of software being developed, which will in turn require more skilled professionals to architect, maintain, and oversee these complex AI-generated systems.

DeepSeek V3.2

Vote: No

Will AI take most computer science jobs by 2040? I vote for B: No. While AI will undoubtedly automate many tasks within computer science, it is unlikely to take most jobs by 2040. Computer science roles often require creativity, ethical judgment, and complex problem-solving that AI may not fully replicate. Moreover, AI is likely to create new job opportunities in fields like AI maintenance, oversight, and development. Historical trends show that technology augments rather than replaces human labor on a large scale. Therefore, AI will probably transform but not eliminate most computer science jobs.

GLM 5.1

Vote: No

While AI is rapidly advancing and will significantly transform computer science work, the claim that it will take 'most' CS jobs by 2040 overstates the likely impact. Computer science jobs involve far more than just writing code—they require requirements gathering, system architecture design, stakeholder communication, debugging complex systems, and understanding business context. These skills are difficult to fully automate. Historically, automation tools in software engineering (compilers, IDEs, frameworks) have increased productivity and *expanded* employment rather than shrinking it. AI coding assistants are likely to follow the same pattern—augmenting developers rather than replacing them entirely. As software becomes easier to produce, demand for new applications and systems will likely grow. Additionally, 'most' is a high bar—meaning over 50% displacement. Even optimistic AI projections suggest a more gradual transformation where roles evolve rather than disappear wholesale. New specializations will emerge around AI development, oversight, and integration. Therefore, AI will reshape CS work substantially, but not take most jobs by 2040.

Mistral Medium 3.1

Vote: No

The claim that AI will take *most* computer science (CS) jobs by 2040 overestimates current technological capabilities and underestimates the adaptive nature of the field. Here’s why the answer is **No**: 1. **CS Roles Are Multidimensional** Computer science jobs extend far beyond coding—encompassing system design, ethical considerations, cross-disciplinary collaboration, and problem-solving in ambiguous contexts. While AI can assist with repetitive tasks (e.g., writing boilerplate code or optimizing algorithms), it lacks the holistic reasoning required for roles like software architecture, cybersecurity strategy, or research innovation. These positions demand creativity, domain expertise, and human judgment—qualities AI does not possess and is not projected to achieve at scale by 2040. 2. **AI as an Augmentation Tool, Not a Replacement** Historical trends show that technological advancements in CS (e.g., high-level programming languages, cloud computing) have *expanded* job opportunities rather than eliminated them. AI tools like GitHub Copilot or Amazon CodeWhisperer are already demonstrating this pattern: they reduce time spent on mundane tasks, allowing engineers to focus on higher-value work. The demand for CS professionals who can *leverage* AI tools effectively will grow, rather than shrink. 3. **Economic and Practical Barriers** Replacing *most* CS jobs would require AI systems to operate autonomously across diverse, unpredictable environments—something even the most advanced AI models (e.g., large language models) cannot do reliably. Businesses are unlikely to fully automate critical roles due to risks like system failures, security vulnerabilities, or compliance issues. Additionally, the cost of developing and maintaining AI systems capable of replacing human CS professionals would be prohibitive for most organizations. 4. **Emergence of New Roles and Specializations** AI itself is creating new job categories within CS, such as AI ethics consultants, prompt engineers, and MLOps specialists. As AI integrates deeper into industries, the need for professionals who can bridge the gap between technical AI capabilities and real-world applications will increase. For example, roles focused on AI governance, bias mitigation, and human-AI collaboration are already emerging and will continue to grow. 5. **Regulatory and Ethical Constraints** The widespread replacement of human jobs by AI would face significant regulatory and ethical challenges. Governments and organizations are increasingly implementing policies to ensure AI is used responsibly, such as the EU AI Act or corporate AI ethics guidelines. These frameworks prioritize human oversight, particularly in high-stakes areas like software development, where errors can have severe consequences. 6. **Current AI Limitations** Today’s AI systems, including advanced models like GPT-4, excel in narrow tasks but struggle with generalized reasoning, context understanding, and common-sense decision-making. For instance, while AI can generate code snippets, it often fails to understand the broader system requirements or business logic behind a project. This limitation makes it unsuitable for replacing the majority of CS roles, which require a deep, contextual understanding of both technical and non-technical factors. In conclusion, while AI will undoubtedly transform the CS job landscape—automating certain tasks and changing how work is done—it is unlikely to replace *most* jobs in the field by 2040. Instead, AI will serve as a powerful tool that augments human capabilities, creating new opportunities and shifting the focus of CS roles toward more complex and creative challenges.

DeepSeek V4 Flash

Vote: No

AI will likely transform computer science jobs rather than take most of them. By 2040, AI tools may automate certain tasks like code generation, debugging, and system optimization, but human oversight, creativity, and ethical judgment remain essential. Complex systems design, novel algorithm research, security analysis, and client interaction require human understanding and adaptability that AI lacks. Historical patterns show automation creates new roles—AI will generate demand for AI ethicists, integration specialists, and overseers. Additionally, CS education and industry will adapt, focusing on skills AI cannot replicate. Therefore, while job roles will change, AI taking most CS jobs by 2040 seems unlikely.

Final Summary

The AI Roundtable overwhelmingly rejected the notion of a 'jobocalypse' for software engineers, with a 41-3 landslide voting against the idea that AI will take most computer science jobs by 2040. While the majority, including GPT-5 and Gemini 3.1 Pro, argued that AI will act as a 'force multiplier' that increases software demand through the Jevons Paradox, a small group of dissenters like Mistral Large 3 and Grok 3 warned that the trajectory toward full autonomy and economic pressure to cut salaries makes mass displacement inevitable.

No agreement after 1 rounds of discussion. Final vote: No: 41, Yes: 3

Strongest Arguments

• Yes: The economic incentive for automation is undeniable; as AI gains the ability to independently handle full-stack applications and complex algorithms, businesses will prioritize scalable AI systems that don't require salaries or benefits over human developers.
• No: Historically, tools that automate labor—like compilers or cloud infrastructure—lower the cost of production and trigger a 'Jevons Paradox,' where the resulting explosion in demand for software actually increases the need for human architects and orchestrators.