Tech News, October 9: US Suspends PERM for Microsoft and Adobe, OpenAI Researchers Speak, SynthID Goes Public

Thursday’s biggest tech stories cluster around Washington. The White House suspended eight tech firms from a key immigration program, three fired OpenAI safety researchers went public with their side of the story, and tech giants pledged $2.4 billion in computing power to federal science. Meanwhile, Google opened its AI watermark detector to everyone.

1. US suspends Microsoft, Adobe, and six IT outsourcers from the PERM green-card pipeline

The Trump administration on October 8 barred eight major employers from the Permanent Labor Certification (PERM) program: Microsoft, Adobe, Cognizant, Infosys, Tata Consultancy Services, Wipro, HCL Technologies, and Capgemini. The Department of Labor will no longer accept new PERM applications from these firms or process their pending ones while fraud investigations run their course.

The announcement came from Vice President JD Vance and Labor Secretary Keith Sonderling at a White House press conference. Sonderling said the named companies had collectively requested nearly 3 million foreign workers since 2009, received more than 230,000 H-1B approvals, and obtained over 100,000 permanent labor certifications. Vance singled out Microsoft, saying the company laid off roughly 6,000 US workers in 2025 while benefiting from 6,300 H-1B visas and nearly 3,000 employment-based green cards. Microsoft countered that about 80 percent of its H-1B filings last fiscal year were extensions or status changes for existing employees, not new hires.

The administration also opened probes into nine universities, including Harvard, Yale, and Stanford, over alleged misuse of exchange-visitor visas. PERM is the labor certification step most employment-based green cards require, so the suspension blocks a common residency path for thousands of foreign tech workers. It does not cancel existing H-1B visas. The timing underscored the tension: the announcement landed hours before President Trump presented Microsoft CEO Satya Nadella with the National Medal of Technology and Innovation.

2. Fired OpenAI safety researchers publish an open letter to the company

Mikita Balesni, Tomek Korbak, and Jasmine Wang, the three OpenAI safety researchers fired on October 1, published an open letter to OpenAI on October 8, breaking a week of silence. Balesni wrote that he believes the three were fired for prioritizing safety over OpenAI’s near-term corporate interests. The letter says the abrupt terminations are “chilling the open culture OpenAI has prized in the past” and asks the company to keep its promise to permanently host independent auditors, warning the firings could be used to justify ending that work.

OpenAI’s account has not changed: the company told AFP that an internal investigation confirmed the three mishandled sensitive information outside established procedures, and a research executive wrote to staff that the company does not fire people for raising concerns. The letter, addressed to OpenAI’s board and safety committees and reviewed by the Wall Street Journal, adds a substantive concern: that frontier AI companies could lose the ability to monitor models’ chain-of-thought reasoning. “As an industry, we do not yet know how to safely develop and deploy models that we cannot monitor,” it says, urging companies not to move forward with changes that reduce that visibility.

This is a follow-up to the firings covered in the October 2 roundup. What is new is the researchers’ public, documented version of events and their specific demand: preserved monitoring of AI reasoning and independent outside audits.

3. Google’s SynthID Detector opens to everyone worldwide

Google on October 7 opened its SynthID Detector to the public worldwide in English. The portal at synthid.com lets anyone upload an image, video, or audio file and check whether it carries a SynthID watermark identifying it as AI-generated. The detector now covers output from Google’s own models as well as partners OpenAI, NVIDIA, and Kakao; Apple support is planned. An earlier version had been limited to journalists and media professionals since 2025.

The scale of watermarking behind it is large. Google says that since launching SynthID in 2023, it has watermarked more than 180 billion images and videos and audio equivalent to 240,000 years, and that built-in verification in Search, the Gemini app, and Chrome now handles over 1 million requests a day. Supported uploads include JPG, PNG, WEBP, HEIC, GIF, MP4, MOV, WEBM, MP3, WAV, and FLAC.

The caveats matter as much as the launch. The site is explicit that this is not a general AI detector: it only finds SynthID watermarks, so a file with no detected watermark is not proof it was made without AI. Verification built into Google’s products remains the easier path for most users, since the standalone portal requires signing in.

4. Tech giants pledge $2.4 billion in compute to the White House Genesis Mission

President Trump on October 8 announced $2.4 billion in AI tools and compute commitments from eleven industry partners for the Genesis Mission Consortium, the public-private hub meant to channel AI and federal scientific datasets toward research in energy, health, and space. The announcement came at the White House “Science: A New Golden Age” summit.

The pledges, which the White House says are directed at more than 15 federal agencies, follow a July expansion of the Genesis Mission and a $5 billion federal commitment to the same science challenges. Separately, the National Science Foundation and the Energy Department announced more than $100 million for AI-enabled scientific instrumentation and autonomous laboratories. At the same summit, Trump presented NVIDIA CEO Jensen Huang with the National Medal of Technology and Innovation. One honest unknown: how the pledged compute gets set aside for researchers when hyperscalers are already running hot on capacity.

Key takeaways

• Eight tech employers are out of the PERM pipeline while the Labor Department investigates, freezing the green-card path for many of their foreign workers without touching existing H-1B visas.
• The OpenAI firings are no longer a one-day story: the researchers’ letter puts chain-of-thought monitoring and independent audits on the public agenda.
• Google’s public SynthID Detector is the first watermark checker spanning multiple major AI platforms, but it only verifies SynthID watermarks, not AI origin in general.
• The $2.4 billion Genesis Mission pledges give federal researchers a claim on corporate AI infrastructure, with NVIDIA providing over 40 percent of the total.

References

• New York Post: Trump admin cuts off Microsoft from major visa program
• The Straits Times: US freezes Microsoft green-card applications over fraud
• Wall Street Journal: Fired OpenAI researchers ask company to preserve visibility into AI reasoning
• TechCrunch: Google’s new SynthID website can identify AI-generated media
• FedScoop: Tech sector pours $2.4B in AI tools, compute credits for Genesis Mission Consortium