Is being a Geotechnical Engineer
at risk from AI?

Not in the foreseeable future. Geotechnical engineering sits at the intersection of physical site work, legal liability, and irreducible subsurface uncertainty. While AI will automate routine calculations and report drafting by 2028, the profession's core value—making defensible recommendations when data is incomplete—requires human judgment and accountability. Professional licensure laws and insurance underwriting further insulate the role. The engineers at risk are those doing purely computational work in back offices; those who visit sites, interface with clients, and own project risk remain essential.

Soil classification from lab data, standard bearing capacity calculations, and boilerplate report sections are already 60-70% automatable with current tools. Slope stability modeling and settlement analysis for simple stratigraphy will follow by 2027. What AI cannot yet touch: site investigation planning, real-time decisions during drilling, constructability reviews, and the judgment calls that come from walking a site and noticing what doesn't match the desktop study. If your day is spent running the same analysis templates on similar sites, that work is vulnerable. If you're solving novel problems or managing field programs, you have years of runway.

Yes, urgently. The competitive advantage goes to engineers who use AI to handle routine work faster, freeing time for high-value judgment and client interaction. Learn to prompt LLMs for report drafting, use AI-assisted parametric modeling to explore design alternatives, and validate AI outputs critically. The goal is not to become a machine learning expert but to control AI as a force multiplier. Engineers who resist will find themselves outbid by peers who deliver faster at lower cost. Start with tools like ChatGPT for technical writing and explore geotechnical-specific AI plugins as they emerge.

Junior engineers face higher near-term risk because their primary tasks—running standard analyses, drafting reports, compiling boring logs—are most automatable. Entry-level hiring may contract as firms use AI to stretch senior capacity. However, juniors who adopt AI early can leapfrog peers by delivering senior-level output volume. Senior engineers are more insulated: their value lies in judgment, client relationships, and signing off on work under their PE license. The risk for seniors is complacency—those who don't adapt to supervising AI-augmented teams will lose ground to competitors who do. The profession will likely see a barbell effect: strong demand for experienced PEs and AI-savvy early-career engineers, with pressure on the middle.

Focus on work that combines physical presence, high consequence, and irreducible uncertainty. Forensic investigation and expert witness work are nearly automation-proof. Difficult ground conditions—expansive clays, karst, seismic liquefaction, mine subsidence—require judgment AI cannot replicate and command premium fees. Construction phase services (observation, testing, change order evaluation) keep you on-site and essential. Avoid becoming purely a report factory for routine residential or light commercial work; that segment will see the most pricing pressure from AI-augmented competitors. If you can, position yourself as the engineer clients call when something unusual or risky appears.

Expect bifurcation. Routine work will see fee compression as AI reduces labor hours required, putting downward pressure on salaries for engineers doing commodity analyses. However, specialists in high-risk projects, difficult sites, and forensic work will see stable or rising compensation due to persistent demand and liability constraints. Geographic factors matter: engineers in major metros with complex projects (seismic zones, dense urban infill) are more insulated than those in markets dominated by simple residential work. Overall, the profession's median salary growth may slow, but top performers who leverage AI and own client relationships will continue to command strong compensation. The key is to be in the latter group.

Adoption is slower than in software or finance due to regulatory inertia, liability concerns, and the profession's conservative culture. As of 2026, most firms use AI sporadically—ChatGPT for writing, some automated data processing—but few have integrated it into core workflows. Expect acceleration over the next 3 years as larger firms build proprietary tools and younger engineers push adoption. The industry's slow pace is actually protective: it gives individual engineers time to adapt and means displacement will be gradual rather than sudden. Use this window to build AI fluency and position yourself as a leader in your firm's adoption efforts.