This Month's Latest Tech News in Eugene, OR - Saturday May 31st 2025 Edition

This month, Eugene and Oregon as a whole are at the epicenter of a pivotal tech policy debate, as state lawmakers advance Senate Bill 686, a first-in-the-nation proposal to require tech giants Google and Meta to pay at least $122 million annually to support local journalism, with 90% of funds allocated based on Oregon newsroom employment and 10% channeled to a new Oregon Civic Information Consortium for rural and underserved outlets.

Advocates highlight the urgent need to address a media landscape where Oregon has lost 75% of its journalism jobs and faces the closure of legacy newspapers, while opponents, including Google and Meta, argue the bill endangers open web access and have threatened to remove Oregon news from their platforms, echoing tactics seen in Canada and Australia.

As reported by The Seattle Times, this legislative effort could reshape public trust and democracy if Oregon succeeds where others have struggled.

In the words of Dr. Courtney Radsch,

“In this pivotal moment for the future of sustainable and independent journalism, it is essential for states like Oregon to lead the fight to protect local news and push back against the monopolization and degradation of our information ecosystem.”

OPB details both the broad bipartisan support and looming legal challenges, while commentary in Oregon Capital Chronicle frames the initiative as a foundation for a more inclusive “civic information infrastructure” - demonstrating how policy, tech, and community action are poised to reshape the region's media future.

Oregon Senate Bill 686 has advanced to a full Senate vote, sparking national attention with its requirement that tech giants like Google and Meta compensate local news providers for distributing Oregon journalism on their platforms.

The bill, modeled after similar legislation in California and Canada, mandates at least $122 million in annual payments - $104 million from Google and $18 million from Meta - to be distributed largely based on the number of journalists employed, with 90% supporting newsroom operations and 10% funding grants via the University of Oregon for industry innovation and training.

Proponents argue SB686 could help reverse the steep decline in local news coverage, restore lost jobs, and ensure robust civic engagement, with the News/Media Alliance stating,

"Local news serves a vital role in communities all over Oregon ... The Oregon Journalism Protection Act would be an important boost in their ability to stay afloat, and provide Oregon communities with the information and local coverage that they deserve."

However, critics from big tech warn the law could trigger news blackouts, echoing Meta's response to similar measures in Canada, and raise constitutional concerns over government-imposed payments for privately created content.

A summary of the bill's key financials is provided below:

Tech Company	Annual Payment
Google	$104 million
Meta	$18 million
Total (Estimated)	$122 million

While legal challenges are expected, supporters see the measure as necessary to break what Dr. Courtney Radsch calls tech firms' "monopolization and degradation of our information ecosystem" (analysis on SB 686).

The bill's progression mirrors global debates about how to equitably sustain local media in the digital age, as discussed in this public radio feature on cross-border media compensation laws.

Oregon State University's College of Engineering has made international headlines this month with the debut of an innovative chip that slashes the energy consumption of large language models (LLMs) like GPT-4 and Gemini by 50%.

This pioneering chip, introduced at the IEEE Custom Integrated Circuits Conference, leverages on-chip AI classifiers to detect and correct data transmission errors far more efficiently than traditional power-hungry methods, dramatically reducing the electricity footprint of data centers - an urgent need as the International Energy Agency projects data center power use will double by 2026 to levels matching Japan's current national consumption.

As doctoral student Ramin Javadi explained,

“We have designed and fabricated a new chip that consumes half the energy compared to traditional designs.”

The research team's work has won the IEEE Best Student Paper Award and attracted support from DARPA and the Semiconductor Research Corporation, setting the stage for even greater energy savings in the next chip iteration.

The implications reach beyond data centers: future consumer devices could also benefit from these advances, making AI-powered technology both more efficient and sustainable.

For a comprehensive technical overview, see Oregon State's official announcement on shrinking LLMs' energy footprint.

Explore the environmental and industry impacts in KPIC's deep dive on OSU's energy-saving chip, and for a global perspective on AI-driven efficiency, consult Securities.io's feature on this and related breakthroughs.

Here's a quick look at the chip's impact compared to industry benchmarks:

Chip Design	Energy Consumption	Error Recovery Method
Traditional (equalizer-based)	Baseline (100%)	Power-hungry equalizers
OSU AI-powered chip	50% less	On-chip AI classifiers

This breakthrough positions Eugene, Oregon, as a leader in sustainable AI infrastructure and signals a major leap toward greener digital innovation worldwide.

This month, Oregon State University's latest research sheds light on how artificial intelligence can be a powerful tool for enhancing student creativity - when paired with strategic instructor guidance.

A recent study demonstrated that students using AI like ChatGPT, but guided by fewer than 20 minutes of targeted instruction, produced notably more creative writing than those who worked solo or used AI unguided.

The study's authors observed that

“unsupported AI use resulted in a ‘flattening' effect: it boosted creativity in lower-performing students but diminished it in highly creative students. When instructor guidance was provided, overall creativity increased significantly, with the greatest improvements among students who initially scored lowest in creativity,”

according to local coverage of the OSU findings.

The experiment, conducted with 31 students in online creative writing courses, used Harvard's consensus creativity-rating method and underscores AI's greatest value: collaboration over replacement.

As co-author Wayne Harrison put it,

“This approach has huge implications, not just for writing classes, but for any discipline where critical or creative thinking matters. We don't have to choose between banning AI entirely and throwing up our hands in defeat.”

Read more about the detailed methodology and the impact on student satisfaction in the original OSU news release, AI Improves Creativity in Student Writing when Supported by Instructor Guidance.

For an in-depth discussion with study co-author J.T. Bushnell, tune into this Jefferson Exchange interview on guided AI and creativity.

Recology King County in Seattle is leading a new wave of sustainable waste management by deploying Glacier's advanced AI robotic sorting technology, which efficiently identifies and sorts more than 30 types of recyclable materials - including PET plastics, aluminum cans, and toothpaste tubes - at a remarkable pace of 45 items per minute.

By integrating four robots and four AI vision systems within its facility, Recology has not only enhanced recovery and purity rates but has also significantly reduced contamination, improving operational safety for its team and producing real-time data insights for greater transparency and value.

These robots, installed for less than half the cost and in a fraction of the space of traditional automation, transformed operations: in a recent example, the system identified 380 tons of lost PET bottles per year and enabled a 70% reduction in material residue, resulting in an additional $138,000 in annual revenue.

The project was fueled by Glacier's recent $16 million funding round and is already part of a nationwide movement to modernize recycling infrastructure in cities from San Francisco to Detroit.

As Recology's CEO Salvatore M. Coniglio explains,

“After seeing Glacier's technology in action at our other facilities, it became clear they offered a faster, safer, and more accurate way to recover valuable materials back into the circular supply chain... excited to bring their AI robotic system to Seattle, one of the nation's most forward-thinking communities when it comes to recycling and sustainability.”

For a detailed look at the data-driven improvements, see Recycling Today's analysis of Recology's Seattle MRF rollout.

Explore the broader impact of Glacier's AI funding and national deployment in this Webwire special report on Glacier's AI funding and deployment, or dive into the technology's local implementation and sustainability goals at KIRO 7's coverage of Glacier robots in King County.

Key Metric	Glacier AI System (Recology Seattle)
Sorting Speed	45 items/minute
Materials Identified	30+ (e.g., PET plastic, aluminum, toothpaste tubes)
Annual Additional Revenue	$138,000 (from PET residue reduction)
Funding Raised	$16 million
Energy Savings (Aluminum)	93% less energy than using primary ore

Duolingo has announced a sweeping “AI-first” transformation, signaling one of the most significant pivots in the edtech industry this year. The company recently launched 148 new language courses - a feat that nearly doubles its previous total, achieved in just one year thanks to generative AI, compared to the 12 years it took to build the initial 100 by hand.

CEO Luis von Ahn emphasized the intention to “gradually stop using contractors to do work that AI can handle” and integrate AI into hiring and performance reviews, aiming to automate repetitive tasks so staff can focus on creative challenges.

As covered in TechCrunch's report on Duolingo's AI-powered course expansion, the shift has spurred rapid scaling but triggered debate over quality and workforce impact.

Notably, according to TechCrunch's analysis of the emerging AI jobs crisis, Duolingo has already offboarded about 10% of its contractors, mostly translators and writers, inviting concerns about job displacement in both creative and entry-level positions.

After initial backlash, Duolingo clarified its stance in Staffing Industry News, stating AI will accelerate - rather than replace - human work, and the company will continue hiring while supporting employees with training and AI tooling.

As this paradigm shift unfolds, both benefits and anxieties remain, exemplified by the table below summarizing key changes at Duolingo:

Initiative	AI-First Impact
Course Creation	148 new courses in one year (vs. 12 years for previous 100)
Workforce	10% contractor reduction; AI to handle repetitive tasks
Hiring & Reviews	AI use included in hiring/performance assessments

"I see [AI] as a tool to accelerate what we do, at the same or better level of quality. And the sooner we learn how to use it - responsibly - the better off we will be in the long run." - Luis von Ahn, CEO, Duolingo

Epic Games' recent acquisition of Loci marks a pivotal moment in AI-driven 3D asset management, promising to drastically simplify how creators organize, search for, and protect digital assets across platforms like Unreal Engine, Fortnite, and the Fab marketplace.

Loci leverages advanced computer vision models - trained on over 1 million 3D assets - to automatically tag and categorize content, addressing long-standing challenges of manual metadata entry that hamper discoverability and workflow efficiency.

Integration with Epic's ecosystem means artists and developers can expect powerful new tools for automated asset indexing, similarity-based search, and IP protection, including enhanced detection of unauthorized content within user-generated projects.

As highlighted in an official statement, Loci's CEO Eugene Yi shared,

“3D is the next frontier of AI-powered creativity, and we believe in making it more accessible, discoverable, and protected. Joining Epic Games allows us to integrate Loci's 3D understanding technology where it matters most - supporting creators directly in the strongest 3D asset ecosystem in the world.”

This acquisition comes amid growing competition in 3D digital asset management, with solutions like echo3D and Sketchfab for Teams also advancing features such as AI tagging and collaboration.

The move signals Epic's commitment to automating artist workflows and establishing industry-leading asset moderation, with broader implications for the entire creative technology sector.

For a detailed look into how Loci's AI technology transforms asset management, see TechCrunch's coverage of the Epic-Loci acquisition, get insights into integration and production pipeline changes at VP Land's analysis of how AI is reshaping asset discovery and protection, and read direct statements from Epic and Loci leaders in the official Epic Games press release on Loci joining their ecosystem.

AI is rapidly transforming the cybersecurity landscape in Eugene, OR and beyond, acting as both a vital defense mechanism and a powerful tool for cybercriminals.

Recent research highlights a surge in sophisticated ransomware attacks, such as the emergence of the AI-driven FunkSec gang, which uses large language models (LLMs) to create highly effective, low-cost attacks targeting sectors like government, finance, and education according to Kaspersky's 2025 Ransomware Report.

Vulnerabilities are also appearing in the software supply chain, where AI-generated code can introduce phantom dependencies - a new risk called “slopsquatting” - that attackers exploit by publishing malicious libraries, posing a growing threat as AI-generated code becomes the industry norm as revealed in Kaspersky's slopsquatting analysis.

Meanwhile, experts caution against relying on AI for critical security tasks like password creation, since AI-generated passwords often contain recognizable patterns that make them susceptible to attacks, with studies showing nearly 88% of passwords from some leading LLMs can be cracked in under an hour.

As Eugene's tech community grapples with these dual realities, cybersecurity leaders emphasize the importance of layered defenses, vigilant code review, and robust user education.

“Supply chain risks may seem overwhelming, but awareness is the first step toward prevention. By testing updates rigorously, leveraging AI-driven anomaly detection, and diversifying providers to reduce single points of failure, we can reduce weak elements and build resilience… human vigilance remains the cornerstone of security.”

Explore more on the evolution of ransomware and essential preventative measures in this in-depth report on AI-driven ransomware threats.

Johns Hopkins researchers are rapidly advancing cancer diagnostics with artificial intelligence (AI), enabling earlier, more accurate detection and improved patient outcomes across multiple cancer types.

Their AI-powered “liquid biopsy” approaches analyze circulating DNA fragments in blood to identify genetic and immune signatures of cancers that were previously hard to detect, such as brain and pancreatic malignancies.

For example, the Leukemia Smart Physician Aid, a cutting-edge AI tool for acute promyelocytic leukemia detection, flags acute promyelocytic leukemia (APL) within hours - potentially saving lives by accelerating treatment for this rare, curable leukemia.

In pancreatic cancer, the ARTEMIS-DELFI technique for early therapy response evaluation evaluates therapy response as soon as four weeks after treatment begins and outperforms traditional imaging, giving clinicians the ability to quickly switch ineffective therapies.

“This algorithm has the potential to add to existing human capacity and not merely substitute for it,” says Dr. Eugene Shenderov, highlighting AI's role in empowering - not replacing - medical professionals.

Meanwhile, next-generation AI assays now detect brain cancer in approximately 75% of cases, compared to under 10% accuracy with previous blood-based tests - a leap that could expedite care for thousands each year.

Learn more about brain cancer screening advances with AI-powered liquid biopsies.

As these breakthroughs attract global attention, sustained research funding remains critical to ensure continued progress and accessibility, especially for rural and underserved communities.

The National Geospatial-Intelligence Agency (NGA) is accelerating responsible AI adoption to safeguard national security in the face of rapidly evolving threats and overwhelming data volumes.

Generative AI is being deployed to empower human analysts, process vast geospatial datasets from satellites and sensors, and maintain decision superiority as adversaries enhance their own AI capabilities.

As an NGA Senior Data Officer stated,

“As our adversaries advance their own AI capabilities, implementing this technology isn't just an option - it's imperative.”

NGA's investments span cutting-edge AI programs - like Project Maven and the five-year ASPEN modernization effort - and the establishment of new leadership positions to govern standards, risk, and ethical deployment practices.

This commitment extends beyond federal agencies, with states enacting executive orders and legislative safeguards to balance innovation and risk management in AI regulation.

The table below summarizes key advancements:

Program	Focus	Highlights
Project Maven	AI/ML in imagery analysis	25,000+ users; 80% reduction in targeting timelines
ASPEN	Automated GEOINT production	Addresses tripling of data; boosts warning accuracy
Generative AI Rollout	AI-created intelligence content	Mitigates human processing gaps with oversight and training

Advanced automation, computer vision, and responsible AI frameworks ensure the fusion of human expertise and AI, leading to faster, more reliable threat detection and disaster response.

For more detail, explore how NGA prepares to use generative AI for mission support, the agency's AI ambitions and resource focus for 2025, and the evolving legal and regulatory landscape shaping responsible AI deployment.

Oregon is taking a substantive lead in responsible artificial intelligence by unveiling a comprehensive state government AI Advisory Council action plan, delivered to Governor Kotek in February 2025.

The plan, built on 74 actionable recommendations, emphasizes the creation of robust AI policies focusing on transparency, effective governance, privacy, and workforce readiness.

Chair Terrence Woods noted,

"The council has delivered a comprehensive AI action plan, aligning with Oregon's core values of diversity, equity, and inclusion."

Oregon's approach stands out for referencing global standards such as the White House AI Bill of Rights, OECD, and EU principles, with the vision to foster public trust and uphold privacy and ethical integrity.

While legislators are not rushing AI-specific laws, the Attorney General's recent guidance reminds businesses that longstanding regulations - including the Unlawful Trade Practices and Consumer Privacy Acts - directly apply to AI. Disclosures must be clear, especially if personal data is used for AI training, and companies must obtain explicit consent for handling sensitive data.

To monitor compliance and progress in this evolving landscape, Oregon will publicly report development metrics within 18 months and provide updates for the upcoming 2027-29 budget.

For key benchmarks from Oregon's privacy and AI efforts, see the table below:

Metric	Value/Status
AI Council Recommendations	74
Consumer Complaints Received (last 6 months)	110
Privacy Law “Cure” Period Expires	Jan 1, 2026

For a detailed analysis of business obligations and enforcement trends under Oregon's privacy law, visit the Oregon Privacy Law six-month update.

Additional AI compliance and regulatory insights are available from the State Attorneys General Guidance on Privacy & AI.

Eugene's tech sector is at a defining crossroads, mirroring global trends as responsible, inclusive AI adoption takes center stage. Oregon's government has moved decisively with the AI Action Plan - a framework emphasizing human oversight, privacy, and workforce development, reflecting a commitment to both innovation and public trust.

As competitive pressures mount - OpenAI reached a $300 billion valuation, and industry giants like Anthropic, Databricks, and xAI secure multibillion-dollar funding rounds according to recent market analyses - Eugene's ecosystem must adapt by fostering ethical leadership and sustainable business models.

Around the state and nationwide, startups are leveraging AI for climate impact, energy savings, and global market advances, highlighted by Neuralix's 30% energy reduction in industrial operations and Microsoft's record purchase of clean cement from Sublime Systems as reported by Greentown Labs.

The convergence of these factors signals both opportunity and challenge - a reminder that building technical skills, ethical sensibilities, and business agility is more crucial than ever.

For local professionals, Nucamp's bootcamps provide a direct path to upskilling in AI, web development, and cybersecurity, empowering Eugene's next wave of innovation to meet the rapidly evolving demands of the digital era.

As Governor Tina Kotek observed,

“We cannot ignore the rapid growth of AI in our lives. It is incumbent on government to ensure new technology is used responsibly, ethically, and securely.”