The Reason Workers Can't Shake Their Anxiety Over the Optimistic View that "AI Will Create a Large Number of Jobs"

Does AI Take Jobs or Create Them?

The biggest debate surrounding AI is shifting from the technology's performance to how it will ultimately change human lives. Writing articles, coding, creating documents, responding to inquiries, generating images and videos—tasks traditionally considered human intellectual labor are rapidly being replaced by AI.

In response to these concerns, NVIDIA CEO Jensen Huang expressed strong optimism. According to a report by TechCrunch, Huang stated at a Milken Institute event that AI is not a threat that will cause mass unemployment but rather a force that will create new jobs on an industrial scale. In his view, AI is not just an evolution of software; it is a "new industrial infrastructure" that involves semiconductors, servers, data centers, power, cooling facilities, robotics, manufacturing, operations, and application development.

The reason this statement attracted attention is not just because the CEO of NVIDIA made a positive remark about AI. NVIDIA is at the center of the current AI boom, having achieved massive growth due to the increased demand for GPUs, AI chips, and AI infrastructure. Therefore, Huang's statement is both a "future prediction by a central figure in the AI industry" and an "industrial vision by the head of a company selling AI infrastructure." As a result, the reception is divided.

Huang's Argument: "Jobs" and "Tasks" Are Different

The important point in Huang's statement is the idea that automating individual tasks with AI is different from eliminating human occupations altogether.

For example, a sales representative's job includes many tasks such as prospect research, email creation, proposal preparation, negotiations, contract discussions, and customer follow-up. AI can streamline tasks like drafting emails, conducting market research, and creating meeting minutes. However, this does not necessarily mean the entire job of sales will disappear. Instead, representatives may be able to spend more time on relationship building, strategic planning, and decision-making with clients, according to Huang's perspective.

This viewpoint aligns with seeing AI not as a "replacement for humans" but as a "tool to amplify human capabilities." In the past, spreadsheet software did not completely eliminate accountants, nor did the internet eliminate jobs in sales and marketing. Instead, it changed the nature of work and created new job categories. Huang believes AI will similarly not only disrupt jobs but also reorganize industries and roles.

However, this explanation requires caution. The view that "tasks may be automated, but jobs will remain" does not apply to all professions. If a large portion of a role consists of tasks that can be replaced by AI, companies may reduce their workforce. Especially in areas like routine document creation, basic research, customer support, data entry, and simple report writing, AI is likely to have a significant impact.

Thus, the issue is not a binary choice of "whether AI takes jobs or not." More accurately, it's about "which tasks are automated, which roles are redesigned, and which individuals can provide new value."

Is AI an Opportunity for "Reindustrialization"?

Huang also positions AI as an opportunity for the reindustrialization of America, which is an important perspective. While AI may appear as a cloud-based service, it relies on physical infrastructure behind the scenes. This includes the semiconductor supply chain for manufacturing GPUs, massive data centers, power grids, cooling facilities, construction, maintenance, networks, and security. To operate AI, real-world facilities and human resources are necessary.

In this context, the jobs created by AI are not limited to engineers and researchers. They also include construction workers involved in building data centers, electricians, plumbers, maintenance personnel, power infrastructure technicians, manufacturing operators, quality control, logistics, and security personnel. As AI adoption progresses, the industrial foundation required to operate AI will also expand. Huang's statement that "AI creates jobs" is backed by this strong demand for physical infrastructure.

Goldman Sachs also analyzes that while some jobs will be affected by automation due to AI, new demand will arise in infrastructure areas supporting the AI boom, such as power and data centers. In other words, while AI may compress some white-collar tasks, it could boost demand in infrastructure, power, construction, semiconductors, and manufacturing.

However, there are challenges here as well. The jobs lost and the jobs created may not occur in the same place, at the same time, or with the same skills. A person who loses a customer support job cannot immediately become a data center electrician. Even if office jobs in urban areas decrease and infrastructure construction demand increases in rural areas, workers may not easily relocate. While AI may increase employment on a macro level, it could cause significant pain on an individual level.

Numbers Indicate Neither Optimism Nor Pessimism, But "Large-Scale Reorganization"

In the AI employment debate, optimism and pessimism often clash. While some, like Huang, claim that "AI creates jobs," others, like Anthropic CEO Dario Amodei, warn that AI will significantly disrupt entry-level white-collar jobs.

BCG points out that a significant portion of jobs in the U.S. will be redesigned by AI, and a certain percentage of jobs may disappear. Goldman Sachs also sees a substantial number of jobs globally being affected by AI automation. The World Economic Forum's Future of Jobs Report 2025 predicts that multiple factors, including AI and information processing technologies, will lead to simultaneous job creation and loss by 2030.

The important point here is that "jobs exposed to AI" and "jobs actually lost" are not the same. Even if AI can automate parts of a job, it doesn't necessarily mean the entire occupation will disappear. Conversely, while the job title may remain, the content may change significantly, and the required skills may become entirely different.

For example, a writer may become someone who verifies information, designs structures, edits AI outputs, and adds unique perspectives, rather than just writing text. A programmer may become someone who defines specifications, evaluates AI-generated code, and handles overall design and quality control, rather than writing code line by line. A marketer may become someone who tests numerous hypotheses using AI and focuses on customer understanding and brand strategy, rather than creating ad copy.

This change is challenging for the job market because AI-optimized workplaces may eliminate the "ladder for inexperienced individuals to grow through simple tasks." Tasks initially handled by young employees, such as preliminary research, meeting minutes, simple document creation, routine coding, and first-line responses, are easily replaced by AI. While this may be efficient for companies, it could result in a loss of training opportunities for young employees.

Reactions on Social Media: Mixed Views of Agreement, Skepticism, and Anxiety

Reactions on social media do not universally accept Huang's optimism. Observations from public searches on LinkedIn, Reddit, and Hacker News reveal three main tendencies.

The first is agreement with Huang. On LinkedIn, opinions emphasizing that AI redefines human jobs rather than replacing humans are prominent. The view is that individuals who can use AI and companies that can integrate AI into their operations will grow, leading to increased employment as a result. Particularly in business-related posts, the idea that "jobs are not taken by AI but by people who use AI" is supported. This is a practical response suggesting that instead of avoiding AI as a threat, one should quickly become adept at using it.

The second is skepticism towards optimism. On Hacker News and Reddit, there are comments pointing out that NVIDIA is a company selling AI infrastructure, and Huang has an economic incentive to strongly affirm AI demand. Even if AI data center construction temporarily creates jobs, there are voices questioning whether this will lead to long-term and stable employment. The demand for construction may decrease once projects are completed, and it may not become sustainable work that absorbs all unemployed individuals.

The third is a middle-ground reaction: "AI may create jobs, but the transition period is the problem." Even if new jobs are created, current workers may not smoothly transition into them. While individuals who can utilize AI are valued, those in jobs easily replaced by AI may face the need for retraining or career changes. On social media, there is more interest in "who can keep up with the changes" and "whether companies and governments will support the transition" than in the benefits of AI itself.

This division in reactions is natural. From the perspective of those who benefit from AI, it appears as a growth engine. From the perspective of those whose jobs are partially taken by AI, it appears as a threat to their livelihood. Both are parts of reality, and looking at only one side does not capture the full picture.

The Real Question for Companies: AI to Reduce People or AI to Enhance People?

The reason companies introduce AI is often to improve productivity. Achieving more results in less time seems compatible with reducing employment. In fact, companies may proceed with hiring freezes or workforce reductions citing AI implementation.

However, productivity improvement does not necessarily lead directly to employment reduction. If a company is in a growth market, efficiency gains can lower costs, allowing for the provision of more products or services, which can result in increased employment. This is the scenario Huang advocates: companies using AI grow, and that growth leads to new hiring.

The issue is how companies use AI. Will they use it as a tool to reduce people and increase short-term profits, or as a tool to expand human capabilities and create new business and customer value? The impact on the labor market can vary greatly depending on management decisions, even with the same AI.

In the short term, AI-driven "labor reduction" is likely to be valued by investors. Reducing labor costs improves profit margins. However, in the long term, there could be side effects such as a break in talent development, loss of on-site knowledge, weakening of organizational culture, and decline in customer service quality. Companies that implement AI merely as a cost-cutting measure and those that do so as a redesign of organizational capabilities will show significant differences in a few years.

How Should Individuals Prepare?

Regardless of whether one believes in Huang's optimism, the direction individuals should take is quite clear. The risk of continuing to work without using AI is increasing.

First, it is necessary to break down one's job into "tasks" rather than "job titles." What tasks in your work is AI good at? Conversely, where are human judgment, responsibility, interpersonal relationships, field understanding, ethical judgment, and creativity required? Without this breakdown, you may misjudge how AI will change your job.

Next, it is necessary to develop the ability not only to produce deliverables using AI but also to evaluate AI outputs. The ability to discern what is correct, what is dangerous, and where human judgment should be added in AI-generated text, code, analysis, images, and proposals is needed. In the AI era, expertise will shift from merely possessing knowledge to being able to verify AI outputs and connect them to real-world decision-making.

Furthermore, the ability to cross job categories will become important. As AI compresses simple tasks, the value of individuals who can cross business understanding, technical understanding, communication, and customer understanding will increase. Merely being able to use AI is not enough; individuals who can determine which challenges AI should be applied to are in demand.

Conclusion: AI Changes the Conditions of Work Before "Taking" Jobs

Huang's claim that "AI creates jobs" is correct in one aspect. New demand is emerging in many areas, including AI infrastructure, semiconductors, data centers, applications, robotics, power, and manufacturing. Companies that master AI are likely to grow and create new job categories.

However, that alone does not eliminate workers' anxieties. Even if AI creates jobs, it may simultaneously reduce some jobs, change job content, and widen skill gaps. While the total number of jobs may be positive, there may be significant disruptions in individual careers.

Therefore, the real question in the AI employment debate is not the simple question of "whether AI takes or increases jobs." The questions to be asked are "whose jobs will change, who can transition to new jobs," "will companies use AI as a tool for workforce reduction or for talent expansion," and "how will society support the pain of the transition period."

AI is not a magic wand that instantly erases human jobs. However, it already has the power to change the criteria for measuring the value of human work. Whether Huang's optimism becomes a reality is not solely determined by the performance of AI itself. It depends on how companies, governments, educational institutions, and working individuals use and prepare for this technology.

Source URLs

TechCrunch: Used to confirm the content of NVIDIA CEO Jensen Huang's claim at a Milken Institute discussion that AI creates a large number of jobs rather than taking them, and the argument of distinguishing between "tasks" and "jobs."
https://techcrunch.com/2026/05/04/as-workers-worry-about-ai-nvidias-jensen-huang-says-ai-is-creating-an-enormous-number-of-jobs/

Milken Institute: Overview of the "Leading in the Age of AI" event where Huang spoke. Used to confirm the discussion theme on the impact of AI on industries, corporate competition, jobs, and human creativity.
https://milkeninstitute.org/content-hub/event-panels/leading-age-ai-conversation-nvidia-ceo-jensen-huang

NVIDIA Official LinkedIn Post: Used to confirm Huang's message that "AI has created over 500,000 jobs" and reactions on LinkedIn.
https://www.linkedin.com/posts/nvidia_the-facts-on-ai-and-the-american-economy-activity-7455728898811645952-splA

BCG: Used to confirm analysis on the reshaping of U.S. jobs by AI and the potential disappearance of a certain percentage of jobs.
https://www.bcg.com/publications/2026/ai-will-reshape-more-jobs-than-it-replaces

Goldman Sachs: Used to confirm analysis on jobs exposed to AI automation, the impact on the U.S. labor market, and employment demand related to data centers and power infrastructure.
https://www.goldmansachs.com/insights/articles/how-will-ai-affect-the-us-labor-market

World Economic Forum: Future of Jobs Report 2025. Used to confirm the context of labor market changes, employment, and skill changes towards 2030.
https://www.weforum.org/publications/the-future-of-jobs-report-2025/

Axios: Used to confirm the warning by Anthropic CEO Dario Amodei on the impact of AI on entry-level white-collar jobs.
https://www.axios.com/2025/05/28/ai-jobs-white-collar-unemployment-anthropic

LinkedIn: Comments on NVIDIA's related posts. Used to confirm SNS reactions that AI creates new jobs but the speed of transition and corporate decision-making are challenges.
https://www.linkedin.com/videos/nvidia_ceo-jensen-huang-joins-rep-ro-khanna-at-activity-7450636478650515457-pwDt

Reddit: Reactions to Huang's AI employment theory. Used to confirm skeptical and discussion-oriented reactions on what AI leaves for humans.