Semiconductor Engineers in the AI Era: At the Crossroads of Job Loss and Responsibility

Up to 300 million jobs could disappear due to AI by 2030.
— Goldman Sachs

AI cannot replace people, but people who use AI... can do more work than many others. Ultimately, many people's jobs could be replaced.

This isn't just my 'personal forecast' — it's from a report by the World Economic Forum (WEF). According to WEF, 40% of employers plan to reduce their workforce due to AI from 2025 to 2030.

• Rule-based office workers → AI Agents
• Taxi drivers and truck drivers → Advances in large-scale transportation technology, drone technology, autonomous driving
• Phone customer service representatives → Online Chatbots

https://www.weforum.org/publications/the-future-of-jobs-report-2025/ The Future of Jobs Report 2025 Learn how global trends like tech innovation and green transition will transform jobs, skills, and workforce strategies in The Future of Jobs Report 2025

Is your livelihood safe from AI?

Looking at Netflix's Black Mirror Series

The general population receives basic necessities like this: 'Underwear (clothing) + meals sustained by pills (food) + cramped rooms with ads playing all day (shelter)'

Among these, people who 'become entertainers... become policy decision-makers... or do something highly productive that AI cannot do' become the upper class.

Otherwise, they struggle to survive by 'making their value lower than AI's and somehow persisting.' Taking less money than what AI/robots cost and doing the work directly.

For example, if AI/robots cost 100 won to clean a bathroom → I'll raise my hand and say I'll do it for 90 won. Whether it's dangerous work or human rights-related work, survival comes first.

The Symbiotic Relationship Between AI and Semiconductors

AI and semiconductors exist in a symbiotic relationship where they need each other. AI requires powerful computational capabilities for large-scale data processing, which is implemented through high-performance semiconductor chips. For example, NVIDIA's GPUs accelerate AI tasks like deep learning and natural language processing.

Additionally, AI is revolutionizing semiconductor design and manufacturing. AI algorithms help with chip layout optimization, defect prediction, and even developing new materials.

Over the past few years, AI has been integrated into many EDA industry products, enabling 'engineers to design more complex circuits with more transistors.'

Statistics

Where Are Semiconductors Heading?

Please refer to the videos below. CEO Jensen Huang addressed this directly:

Semiconductors are the highest-cost component in 'Consumer Electronics Products.'

For example, the energy consumption of AI chips can place a significant burden on the environment. Training large language models results in hundreds of tons of carbon emissions, and as simulations and inference accumulate, enormous carbon and water consumption will occur.

Can this really be 'a way the Earth can sustain'? This threatens sustainability. To address this, we need to develop power-efficient chip designs and Green AI technologies.

Moreover, AI's ethical issues cannot be overlooked. If AI-designed chips fail or produce biased results, who bears the responsibility?

New Opportunities and Adaptation

While AI threatens jobs, it simultaneously creates new opportunities. New roles like AI ethics specialists and edge AI developers are emerging.

In the semiconductor design field, it seems like diversification will open up through:

• Advanced process nodes + 3D ICs + customized semiconductors

AI EDA tools show high productivity, and a few engineers will control them... but the semiconductors themselves that need to be designed will also diversify, and many new technologies are expected to emerge.

In any case, we'll need to invest in continuous learning and retraining for new things. I'm wondering what would be good to learn.

Final Thoughts

As semiconductor engineers, we stand at a unique crossroads. We're simultaneously the architects of the AI revolution and potentially its casualties. The chips we design today power the AI systems that might replace us tomorrow.

But perhaps that's not the right way to think about it. Instead of fearing replacement, we should focus on evolution. The question isn't whether AI will change our field — it's how we'll adapt and grow with it.

The future belongs to those who can bridge the gap between human creativity and AI capability. In semiconductor engineering, this means understanding not just how to design chips, but how to design chips that serve humanity's best interests while pushing the boundaries of what's possible.

What skills do you think will be most valuable for semiconductor engineers in the AI era?

The intersection of responsibility and opportunity has never been more critical in our field.