Skip to content
The Crash Log
AI & Tech Gone Off the Rails
Fund
Cover image for The Crash Log newsletter
Issue #011 · April 3, 2026

Training Your Replacement

OpenAI pays workers to teach AI their jobs, Dorsey declares managers obsolete, and Perplexity gets caught sharing your private conversations.

RUNTIME_ERROR

OpenAI Is Paying Thousands of Freelancers to Train ChatGPT to Do Their Jobs

A new OpenAI initiative, “Project Stagecraft,” is paying between 3,000 and 4,000 freelancers at least $50 per hour — and up to $500 for domain experts — to build occupation-specific training data for ChatGPT across more than 400 distinct job titles (Source: Business Insider). The project runs through data-labeling startup Handshake AI, with contractors spanning fields from commercial aviation and emergency medicine to plant science, pharmacology, and agricultural management (Source: Startup Fortune).

Contractors are directed to develop detailed personas and simulate real workflows, providing "context, goals, references, and deliverables" to help train the models with specialized human expertise. A Handshake training guide reviewed by reporters indicates the focus is on "knowledge work, not manual labor," with the data collected intended to "map economically relevant tasks and evaluate the model's capabilities."

One contractor who participated told reporters, "We all were aware that we were basically training AI to replace us" (Source: Business Insider).

The revelation arrives as OpenAI has simultaneously begun drafting policy papers on economic disruption and "rethinking the social contract" in an age of AI-driven labor displacement (Source: Vanity Fair).

The company recently closed a record $122 billion funding round at an $852 billion valuation, with enterprise revenue now accounting for over 40% of its $2 billion-per-month run rate (Source: OpenAI).

Stagecraft suggests that OpenAI's ambitions extend well beyond code generation into a systematic cataloging of professional knowledge — field by field, task by task.

DEPRECATED

After Cutting 4,000 Workers, Dorsey Says AI Should Replace Every Middle Manager

Jack Dorsey, the co-founder of Twitter and CEO of Block, published a blog post co-authored with Sequoia Capital partner Roelof Botha arguing that AI can and should replace the entire layer of middle management across corporate America (Source: Block). The post frames Block's recent decision to cut more than 4,000 employees — over 40% of its workforce — not as a cost reduction but as a permanent structural bet on AI-powered operations (Source: CoinDesk).

Dorsey's thesis is that corporate hierarchy exists to solve one problem — routing information through organizations too large for a single person to oversee — and that AI has made the entire structure obsolete.

Block now assigns every remaining employee one of three roles: builders, problem-owners over specific outcomes, and player-coaches who develop talent. An AI-driven "world model" aggregates internal data from code, decisions, workflows, and performance metrics to create a continuously updated picture of company operations, replacing the context that managers traditionally carried (Source: Bloomberg).

Dorsey has said the restructuring was triggered by a capability shift he observed in December in tools including Anthropic's Opus 4.6 and OpenAI's Codex 5.3, which he said were now capable of operating effectively in large codebases (Source: CNN). Block is fully remote, and Dorsey argues that every decision, design, and plan already exists as a digital record — giving AI the raw material it needs to render the managerial layer redundant.

The post arrives as LinkedIn data shows job postings with "manager" in the title have dropped 12% year-over-year in early 2026 (Source: Block).

ACCESS_DENIED

Perplexity AI Sued for Secretly Funneling User Conversations to Meta, Google

A class-action lawsuit filed on April 1 in the U.S. District Court for the Northern District of California accuses Perplexity AI of embedding "undetectable" tracking scripts into its search engine that automatically transmit users' conversations to Meta and Google in real time (Source: Bloomberg). The complaint, Doe v. Perplexity AI Inc., alleges that the trackers are installed the moment users log in, giving Meta and Google full access to every conversation conducted through Perplexity's AI search engine.

The plaintiff, identified only as John Doe from Utah, says he shared personal information about his taxes, investments, and family finances with Perplexity, believing those conversations were private.

According to the complaint, the data sharing persists even when users activate Perplexity's "Incognito" mode, which the company markets as a privacy-first feature.

The tracking allows Meta and Google "to exploit this sensitive data for their own benefit, including targeting individuals with advertising and reselling their sensitive data to additional third parties" (Source: Benzinga).

A Perplexity spokesperson stated, "We have not been served any lawsuit that matches this description, so we are unable to verify its existence or claims."

The lawsuit also names Meta and Google as defendants, accusing them of violating federal and state computer privacy and fraud laws by receiving and exploiting the data (Source: Analytics Insight).

FATAL_EXCEPTION

Iran's Revolutionary Guard Names 18 U.S. Tech Companies as Military Targets

Iran's Islamic Revolutionary Guard Corps (IRGC) declared 18 U.S. technology companies "legitimate targets" on Tuesday, warning that their infrastructure across the Middle East would face attacks beginning Wednesday, April 1, at 8 p.m. Tehran time (Source: The Hill).

The named companies include Apple, Microsoft, Google, Nvidia, Meta, Cisco, HP, Intel, Oracle, IBM, Dell, Palantir, JPMorgan, Tesla, GE, Boeing, Spire Solutions, and UAE-based AI firm G42 (Source: CNBC).

In a statement posted to a Guard-affiliated Telegram channel, the IRGC declared that American AI and information technology companies are "the main element" in designing and tracking the "terrorist operations" conducted by the United States against Iran. The Guard warned employees at the named companies to leave their workplaces immediately "to protect their lives." The threats focus on company infrastructure in the Middle East, not U.S. domestic operations (Source: Time).

The declaration represents an escalation in the ongoing Iran-U.S. conflict, extending the battlefield to the corporate infrastructure that powers AI development, cloud computing, and defense technology across the Gulf region. The IRGC's message stated: "From now on, for every assassination, an American company will be destroyed."

The threats come as the UAE has begun revoking Iranian visas and is preparing to help the U.S. and allies open the Strait of Hormuz by force (Source: CNBC).

Stack Trace

Maine is set to become the first U.S. state to ban new large-scale data centers. The legislation (LD 307) would pause construction of facilities with electric loads of 20 megawatts or more until roughly October 2027, giving a new "Data Center Coordination Council" time to study environmental and economic impacts (Source: The Wall Street Journal). Several proposed projects, including a $5 billion facility in Wiscasset and a $300 million center in Lewiston, have already fizzled amid residential pushback over electricity costs, which are among the highest in the nation (Source: Bangor Daily News).

A new Quinnipiac University poll finds that AI usage among Americans jumped 14%, but trust, optimism, and job confidence all moved in the opposite direction. Seventy percent of respondents now believe AI will shrink overall job opportunities, a 14-point increase, with Gen Z the most pessimistic at 81%. Only 5% believe AI is being developed by people who represent their interests, and 74% say the government is not doing enough to regulate it (Source: Quinnipiac University). Meanwhile, 15% of Americans say they would be willing to work under an AI supervisor — a small but telling signal as companies like Amazon and Block actively dismantle their management layers (Source: TechCrunch).

An estimated 50,000 people and more than 70 organizations marched in San Juan on Saturday, March 28, to protest a $2 billion luxury megaproject backed by foreign capital that would transform 1,500 acres of ecologically sensitive coastline in Cabo Rojo, Puerto Rico, into a private enclave of multimillion-dollar residences, golf courses, and a private airstrip. The project, Esencia, has received nearly $498 million in tourism tax exemptions while the University of Puerto Rico — the island's largest public university — has lost a comparable amount to austerity cuts (Source: The Latino Newsletter). Construction would demand 1.25 million gallons of water per day — roughly a quarter of Cabo Rojo's current consumption — in a region already facing chronic water shortages (Source: Centro de Periodismo Investigativo).

Don't miss the next issue

Subscribe