
Will AI Replace Engineers?
Chip-design agents and CAD assistants are real, and big engineering firms are rolling out AI tools. The evidence so far shows AI changing engineering tasks while licensure, liability and weak benchmark results keep engineers in charge.
The fear is simple: software that drafts designs, writes chip logic and runs checks will make engineers optional. The evidence points somewhere less dramatic.
The BLS projects engineering jobs will grow from 1,835,700 in 2025 to 1,991,000 in 2035. That is a +8.5% change, compared with 3.5% for all occupations. The BLS expects 106,500 openings a year, and the median annual wage in 2025 was $108,620.
Exposure studies tell a similar story. An Anthropic study, summarized by Monograph, put architecture and engineering at about 5% observed exposure against 85% theoretical. It found no systematic increase in unemployment for architects and engineers since AI tools became widely available in late 2022. Dezeen's coverage described the observed exposure as quite low, close to zero.
In other words, AI could touch a lot of engineering tasks on paper. So far it has touched few of them in practice.
Where AI already sits in engineering workflows
Software vendors and the largest design firms are moving fastest. The evidence documents these cases:
- Siemens, 2024-11-13: Siemens announced natural-language assistance for its NX X CAD software. According to Leo AI's tool comparison, the assistant recommends solutions based on best practices and generates automation code.
- Synopsys, March 11, 2026: Synopsys unveiled AgentEngineer spec-to-RTL, which covers specs, RTL, testbenches and verification, Tom's Hardware reports. The same report says Cadence, Synopsys and Siemens all offer agentic AI for chip design, and their claims of autonomy vary.
- Synopsys and NVIDIA, 2026: At the 2026 DAC Chips to Systems Conference, Synopsys showed agentic AI capabilities it built with NVIDIA, EE Times Asia reports. Synopsys also introduced long-horizon AI agents for systems engineering.
- Synopsys productivity claims: In its AgentEngineer webinar, Synopsys cites early customer results of up to a 2X productivity boost, with select cases reaching 5X. These are the vendor's own numbers.
- WSP, 2026: On its Q1 2026 earnings call, WSP's leadership framed AI as a way of expanding effective labor supply rather than replacing engineers, according to a construction earnings review. WSP UK's head of AI, Dan Scott, and head of digital advisory, James Montrose, described to New Civil Engineer how the firm uses AI to save engineers' time.
- AECOM, 2026: AECOM pointed to its acquisition of the AI startup Consigli and said it spent $13 million on its AI roadmap, per the same review.
Job cuts linked to AI are fewer and harder to read. Thomson Reuters announced plans to eliminate up to 500 engineering positions. It also plans to hire over 250 senior and AI-native engineering roles over the next two years. Dow announced 4,500 job cuts in January 2026 as its focus shifts to AI, but the source does not say how many of those jobs were engineers. Our evidence contains no documented case of civil, mechanical or electrical engineers being laid off with AI named as the cause.
Engineering skills gaining value as AI spreads
Responsible charge
ASCE states: "AI cannot serve as a replacement for the professional judgement of a licensed Professional Engineer." A legal analysis notes that professional councils treat AI-assisted work as part of the existing responsible-charge and sealing duty. That makes the stamp more valuable, not less.
Validation and verification
Synopsys says AI-generated results must undergo rigorous engineering validation, ENGtechnica reports. Engineers who can catch a plausible but wrong output will be the ones running these tools.
Constraint setting and data quality
With generative design, engineers specify constraints and the system produces dozens of buildable options in hours instead of weeks, says a tool review. The output depends on the inputs. In a survey by Avnet, almost half of engineers (46%) cited data quality issues as a top challenge when integrating AI into design, per Design News.
AI oversight for public safety
The NSPE position adopted in September 2023 holds that people who oversee AI systems with a direct impact on public safety should meet engineering-licensure standards. That points to a new kind of engineering role.
The messy middle: benchmarks, liability and hype
The benchmarks are humbling. On the EngDesign test, models still failed Analog IC design tasks even after ten iterations. In a Research in Engineering Design study, every model failed every attempt on questions D08 and D10. The EngiBench authors conclude that current LLMs are still far from reliably solving real-world engineering problems.
Liability is real. ASCE's magazine warns that overreliance on AI may entice an engineer to exceed their technical competence. In an NSPE ethics case, the board found that "Engineer A did not maintain responsible charge in violation of licensure law." The Texas engineering board also took up AI policy for licensees in 2024.
Open questions remain. The BLS Monthly Labor Review says aerospace engineers may see some specific tasks completed or aided by GenAI. An Emerald study of civil and environmental engineering argues that institutional resistance shapes real exposure, not technical capability alone.
What to do next, by who you are
- Student: Aim for licensure. The professional judgement that ASCE says AI cannot replace is the part of this job that holds its value.
- Mid-career engineer: Learn the AI features already in your CAD or EDA tools, and build a reputation as the person who checks their output.
- Employer or client: Treat vendor speedup claims as claims, fix your data first, and keep a named engineer in responsible charge of every AI-assisted deliverable.
Myths vs reality for engineers
| Myth | Reality |
|---|---|
| AI is already wiping out engineering jobs. | The Anthropic study found about 5% observed exposure and no systematic rise in unemployment for engineers (Monograph). |
| Chip-design agents run without people. | Synopsys says "human engineers will and must always be in the loop" (Tom's Hardware). |
| AI output can carry its own sign-off. | ASCE says AI cannot replace a licensed PE's professional judgement (ASCE). |
| Demand for engineers is shrinking. | Top design firms saw a 7.4% revenue increase to $158.7 billion (ENR), and WSP says the global engineer pool grows by just 1% (earnings review). |
Our verdict: AI reshapes engineering, it does not replace engineers
We think AI will take over more drafting, code generation and verification grunt work, especially in chip design. Our read is that licensure law, weak benchmark results and strong demand keep engineers in charge for the foreseeable future. Engineers who refuse to learn these tools face the biggest risk.
Questions people ask
Will AI replace engineers?
Not on current evidence. An Anthropic study found about 5% observed exposure for architecture and engineering and no systematic increase in their unemployment since late 2022. The BLS projects engineering jobs will grow +8.5% between 2025 and 2035.
Can AI sign off on engineering drawings?
No. ASCE states that AI cannot serve as a replacement for the professional judgement of a licensed Professional Engineer. Professional councils treat AI-assisted work as part of the existing responsible-charge and sealing duty.
Which engineering fields are seeing AI first?
Chip design leads. Cadence, Synopsys and Siemens all offer agentic AI for chip design, and Synopsys unveiled AgentEngineer spec-to-RTL on March 11, 2026. The BLS also notes that aerospace engineers may see some tasks completed or aided by GenAI.
How good is AI at real engineering design problems?
Still weak. On one benchmark, models failed Analog IC design tasks even after ten iterations. Another study found every model failed every attempt on two design questions.
Sources
Every number, quote and example on this page was checked by script against the sources below before publishing: quotes must match the source word for word, and figures must appear in a source or in BLS data. Our verdict is editorial judgment. Spotted an error? Tell us.
- U.S. Bureau of Labor Statistics: Occupational projections, 2025-35 (Table 1.2)
- Best AI for Engineering: 5 Tools Compared by Job - Leo AI - Generative AI for Engineering CAD Design (getleo.ai)
- AI Integration Drives Engineering Product Growth (designnews.com)
- Best AI for Engineering Design: 5 Tools Tested in 2026 (dancumberlandlabs.com)
- Artificial Intelligence | National Society of Professional Engineers (nspe.org)
- Use of Artificial Intelligence in Engineering Practice | National Society of Professional Engineers (nspe.org)
- Beyond technical capability: how institutional resistance shapes generative AI exposure in civil and environmental engineering occupations (emerald.com)
- Architects and engineers among professions most automatable by AI according to Anthropic (dezeen.com)
- Why Architecture and Engineering Have the Most Untapped AI Potential of Any Industry | Monograph (monograph.com)
- Incorporating AI impacts in BLS employment projections: occupational case studies : Monthly Labor Review : U.S. Bureau of Labor Statistics (bls.gov)
- What the Construction Giants are saying about Tech | Q1 2026 Earnings Report Review (contechroundup.substack.com)
- Interview: WSP experts on how AI is being used to save engineers’ time (newcivilengineer.com)
- ENR 2026 Top 500 Design Firms (enr.com)
- Synopsys, NVIDIA Advance Autonomous AI Agents for Chip Design and Engineering - EE Times Asia (eetasia.com)
- Meet Synopsys AgentEngineer: Your AI Coworker | Synopsys (synopsys.com)
- The state of agentic AI in chip design tools in 2026 (tomshardware.com)
- Synopsys Outlines Three Paths to Agentic AI Adoption in Chip Design - ENGtechnica (engtechnica.com)
- Synopsys Introduces AI Agents for Systems Engineering - ENGtechnica (engtechnica.com)
- AI and Your Stamp: What Architects and Engineers Still Owe | TLY (theleveragedyears.com)
- pac minutes nov 2024 (pels.texas.gov)
- Toward Engineering AGI: Benchmarking the Engineering Design Capabilities of LLMs (arxiv.org)
- A Benchmark for Evaluating Large Language Models on ... (aclanthology.org)
- Evaluating large language models on text-based engineering design tasks: performance gaps and failure modes - Research in Engineering Design (link.springer.com)
- Policy statement 573 - Artificial intelligence and engineering responsibility (asce.org)
- Mishandling AI tools puts civil engineers at risk for ethical violations (asce.org)
- List of Companies Announcing AI-Driven Layoffs - Programs.com (programs.com)
- Thomson Reuters to Slash Up to 500 Engineering Jobs as AI Reshapes Workforce — BigGo Finance (finance.biggo.com)