🔍 Read the full analysis: AI And The Changing Nature Of Work: November 19, 2026 on ThorstenMeyerAI.com
Get audio and creator gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
ThorstenMeyerAI.com’s Nov. 19 report connects four developments affecting work: Rockstar employees are voting on union recognition, evidence points to pressure on early-career software workers, Anthropic disclosed a large financing and infrastructure needs, and security teams faced a compromised package and a Java vulnerability. The report’s interpretation is that workers’ input, training and security responsibilities are becoming central to AI adoption; several outcomes and longer-term effects remain unresolved.
Workers’ influence over AI is emerging across industries, according to a Nov. 19 report from ThorstenMeyerAI.com, which connects a union vote at Rockstar Games with evidence about entry-level software work, Anthropic’s financing and two security alerts. The developments show different pressures—from bargaining over workplace policies to maintaining the systems and infrastructure AI depends on—but do not establish one uniform effect of AI on jobs.
More than 1,600 Rockstar UK employees are eligible to vote on recognizing the Rockstar Game Workers Union, part of the IWGB. The ballot closes Nov. 26. The Register reported that an agreement would allow negotiations over pay, overtime, pensions, bonuses and leave, and provide consultation rights on HR policies and AI. The result and full agreement are not yet known.
In the US, 4.5% of software development job postings on Indeed were entry-level in the first quarter of 2026, compared with 69.3% senior postings; across all US postings, about 46% were entry-level. Indeed’s software development postings were up about 23% year over year in November. Separately, Stanford researchers described employment among 22- to 25-year-olds in AI-exposed occupations as 19% below the level it would have reached if it had kept pace with less-exposed peers. They characterized the result as an early, descriptive indicator, not a causal estimate.
The report also describes an Anthropic randomized experiment involving 52 mostly junior engineers learning a Python library. Participants using an AI assistant scored 50% on a subsequent quiz, versus 67% for those coding by hand; the largest gap was in debugging. Those who asked the assistant for explanations scored 65% or higher. The publisher disclosed that it researches and drafts with Anthropic’s Claude, a potential conflict of interest for its coverage of the company.
Worker Input Meets AI Adoption
Together, the examples point to different kinds of work being reshaped, rather than a single story of jobs disappearing. At Rockstar, workers are seeking a formal channel to discuss AI alongside hours and pay. In software, the available posting figures suggest fewer entry-level roles than senior ones, while overall demand for software developers has not vanished. The experiment raises a separate question: whether relying on AI to produce code can leave new engineers less prepared to inspect and repair it.
Security and infrastructure add another layer. AI services and agents depend on software packages, credentials, computing capacity, electricity and specialized labor. A compromised package or unpatched vulnerability can create work for security teams, while investment in data centers and energy projects can increase demand for technical and construction skills. These are the report’s labor-market interpretations, not proof that every AI deployment will have the same effects.
For employees and managers, the immediate stakes include who has a voice in deployment, whether junior workers can develop the skills needed for later roles, and who is responsible for security upkeep. Those choices can affect workplace conditions as well as whether organizations can safely use AI tools.
Products Worth Considering

LG UltraGear 27-inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDR True Black 400

Bose QuietComfort Ultra Earbuds (2nd Gen), Wireless Noise-Cancelling Earbuds, Black

Bose QuietComfort Wireless Noise Cancelling Headphones – Twilight Blue Limited Edition
As an affiliate, we earn on qualifying purchases.
Four Developments Behind the Report
Rockstar’s vote coincides with the global, local-midnight rollout of Grand Theft Auto VI on Nov. 19. AFP reported that Rockstar has 12 studios and more than 4,700 employees, and that development lasted more than eight years. Take-Two chief executive Strauss Zelnick said in February that generative AI “has zero part” in the game. On Nov. 9, he told investors, “Hit creation remains the purview of human beings,” while saying recent advances had made it easier to create assets. Rockstar publishing head Jennifer Kolbe told Bloomberg, as quoted by The Register, “if they want a union, they will have a union.”
The financing story concerns a reported $107.3 billion deal involving Anthropic, including a $26 billion private placement, according to Bloomberg and Reuters. The company’s filing lists computing and infrastructure investment, as well as hiring and retaining highly skilled staff, among the planned uses. It reports compute and infrastructure commitments of about $900.9 billion, with roughly $451.6 billion before the end of 2030. The filing also warns that rapid job displacement could prompt governments to slow AI adoption and says there is no assurance the company’s efforts to ease displacement will work.
The security events happened on Nov. 17. Versions 1.26.1 and 1.26.2 of Microsoft’s Agent Framework Python package carried a credential stealer, according to the report; PyPI quarantined them about 31 minutes and eight minutes after their respective uploads. CISA separately urged organizations to patch Java4Shell (CVE-2026-114938). Its record listed no exploitation, and the vulnerability requires an x86-64 processor with AVX-512 VBMI.
““Hit creation remains the purview of human beings.””
— Take-Two chief executive Strauss Zelnick, speaking to investors on Nov. 9
Votes, Jobs and Risks Still Unsettled
The Rockstar ballot closes Nov. 26, and its result, the complete terms of any agreement and a tribunal judgment concerning 2025 dismissals were not available in the report. The published figures also do not establish how many jobs AI itself has caused to change or disappear. Stanford’s researchers explicitly caution that their findings are descriptive, and job-posting shares do not by themselves show why employers are advertising fewer entry-level roles.
The Anthropic financing figures describe a deal and commitments, not necessarily completed spending or realized hiring. The report does not establish how many jobs the investment will create, how many roles AI may displace, or whether the company’s measures will offset displacement. In security, the supplied account does not say how many environments installed the affected package versions or whether credentials were misused. CISA’s record listed no exploitation for Java4Shell at the time described; that status could change.
Ballot Results and Security Response
The Rockstar union ballot closes Nov. 26, after which employees and the company will know whether recognition has been approved. Further detail on any agreement and the reported tribunal judgment could clarify how workplace consultation and AI policies may be addressed.
Organizations that installed either affected Agent Framework package version are advised, according to the GitHub Advisory Database guidance cited in the report, to treat the environment as exposed, rebuild it and rotate every reachable credential, including CI and GitHub Actions secrets. CISA’s instruction is to patch Java4Shell. The longer-term labor picture will depend on future hiring data, training practices, investment and how employers negotiate the use of AI.
Key Questions
What is Rockstar employees’ union vote about?
More than 1,600 UK employees can vote on recognizing the Rockstar Game Workers Union, part of the IWGB. The ballot closes Nov. 26. The reported agreement would cover negotiations on pay, overtime, pensions, bonuses and leave, and consultation on HR policies and AI.
Does the coding evidence show that AI is eliminating software jobs?
No. The report cites a low share of entry-level software development postings and a Stanford comparison involving young workers in AI-exposed occupations, but the researchers called their result descriptive rather than causal. Indeed also reported that US software development postings were up about 23% year over year in November 2026.
What did the coding experiment find?
In a randomized experiment with 52 mostly junior engineers learning a Python library, those using an AI assistant scored 50% on a quiz, compared with 67% for those coding by hand. Participants who asked for explanations scored 65% or higher; the largest gap was in debugging.
What should organizations do about the compromised package versions?
The cited GitHub Advisory Database guidance says to treat any environment that installed versions 1.26.1 or 1.26.2 as exposed, rebuild it and rotate credentials it could reach, including CI and GitHub Actions secrets. The report does not say how many organizations were affected.
Has Java4Shell been exploited?
The CISA record cited in the report listed “Exploitation: none” at the time. CISA urged organizations to patch the vulnerability, which requires an x86-64 processor with AVX-512 VBMI.
Source: ThorstenMeyerAI.com
