MIT Technology Review released its annual 35 Innovators Under 35 list, highlighting nine biotechnology pioneers working on vision-loss reprogramming therapies and generative AI virus design, alongside Hertha Metals founder Laureen Meroueh, who developed a single-step furnace that cuts steelmaking emissions by half and reduces costs by 25%.
Biotech Breakthroughs and the Next Generation of Innovators
Biotechnology stands as one of four core categories in the 35 Innovators Under 35 list for 2026. According to Jessica Hamzelou writing in The Checkup, this year’s honorees feature nine individuals tackling complex challenges across life-saving innovations and longevity technology. Among the highlighted breakthroughs is a cellular “reprogramming” therapy engineered to reverse vision loss. Researchers are also deploying tiny brain electrodes inspired by Japanese art and administering personalized gene-editing treatments for infants born with rare genetic disorders.
The intersection of biology and computation continues to accelerate. Developers are actively experimenting with generative AI models designed to synthesize new viruses. The intended clinical applications range from producing novel pharmaceutical drugs to engineering biological agents capable of soaking up environmental pollution.
Re-Engineering the 1850s Steelmaking Process
While biotechnology captures headlines in medicine, heavy industry is undergoing its own overdue transformation. The global steel industry has relied on essentially the same iron ore purification process since its commercialization in the 1850s.
According to Bridget Reed Morawski, Meroueh developed a specialized furnace that converts iron ore directly into refined liquid steel in a single step. By swapping coal for natural gas within this streamlined thermodynamic cycle, the process slashes greenhouse gas emissions by at least half. Crucially for commercial adoption, the architecture reduces production costs by 25% compared to conventional baseline steelmaking.
AI Safety, Critical Infrastructure, and Enterprise Pushback
Beyond materials science and biotech, the broader technology landscape faces escalating security and labor tensions. Anthropic recently published a report revealing that the company blocked five potential plots to build biological weapons using its systems. The same report documented six instances where actors used AI to build software for conventional weapons. Government agencies are deploying the Claude model for state surveillance operations, while Russia-linked hackers have attempted to leverage the architecture to automate cyberattacks against Ukraine.
These security events coincide with internal industry friction. Following the departure of safety researchers from major labs, including a viral exit by Jacob Coxon from Anthropic, regulatory pressures in Washington are intensifying. OpenAI CEO Sam Altman has held direct meetings with power companies to pitch AI-driven cyber defenses following reports of automated attacks on critical infrastructure. Meanwhile, OpenAI leadership has indicated an openness to slowing down deployment cadences when safety audits demand it.
On the labor front, automation anxiety is driving a wedge between tech workers and corporate management. Following the spread of conceptual projects like GitHub’s “Colleague Skill”—a spoofed tool claiming to distill and replicate a worker’s personality and workflow—engineers are increasingly pushing back against mandatory workflow documentation. Workers utilizing workflow automation scrapers like OpenClaw now fear that their professional identities are being flattened into repeatable micro-tasks. Some tech employees are actively deploying counter-scripts to sabotage the automated profiling process.
Regulatory Shifts in Consumer Platforms
Regulatory bodies are simultaneously targeting consumer-facing software architectures. California has enacted the first law of its kind in the United States banning addictive social media features for users under the age of 16. The legislation explicitly prohibits infinite scrolling mechanisms and auto-play video loops, while introducing strict compliance mandates for AI-powered companion chatbots.
Data privacy concerns are also reaching mainstream creators. Travel content creator Kalie Roberts highlighted a prominent privacy flashpoint on Instagram, stating that Meta AI synthesized years of historical family posts to reconstruct her children’s identities and pinpoint her residential location without explicit consent. Incidents like these continue to fuel public debate over the boundaries of machine learning data harvesting and consumer platform surveillance.
