The Human Cost of the AI Buildout

On Tuesday, Oracle began notifying up to 30,000 employees across the United States, India, Canada, and Mexico that their jobs had been eliminated, CNBC reported. Workers received termination emails from "Oracle Leadership" at approximately 6 a.m. local time with no prior warning from HR or direct managers.

"After careful consideration of Oracle's current business needs, we have made the decision to eliminate your role as part of a broader organizational change," the email read. "As a result, today is your last working day." Access to company systems was cut immediately (Source: Rolling Out).

The cuts represent roughly 18% of Oracle's approximately 162,000-person global workforce, making it the largest layoff in the company's history (Source: TNW). India alone lost an estimated 12,000 positions. Senior manager Michael Shepherd confirmed that the cuts hit engineers and leaders alike and were not performance-based (Source: Storyboard 18).

The layoffs are directly tied to Oracle's aggressive expansion into AI infrastructure. According to TD Cowen analysts, the job cuts are expected to free up between $8 billion and $10 billion in cash flow to fund a buildout estimated at $156 billion in total capital spending. Oracle disclosed a $2.1 billion restructuring plan in a recent SEC filing, with nearly $1 billion already recorded (Source: Washington Times).

Anthropic accidentally exposed the full source code of its AI coding tool Claude Code through a routine npm package update on Tuesday, CNBC reported. A misplaced source map file in version 2.1.88 pointed directly to a zip archive on Anthropic's cloud storage containing nearly 2,000 files and over 500,000 lines of internal TypeScript code.

Security researcher Chaofan Shou spotted the exposure and posted about it on X, where it racked up over 28 million views within hours.

The leaked code revealed 44 feature flags for unreleased capabilities, including a persistent background agent called KAIROS, cross-session memory architecture, and a deep-planning system (Source: VentureBeat). Internal codenames also surfaced, including "Capybara," mapping to a Claude 4.6 variant already at version 8. The code was quickly mirrored on GitHub, where it surpassed 84,000 stars and 82,000 forks, making containment effectively impossible (Source: Fortune).

Anthropic called the incident "a release packaging issue caused by human error, not a security breach" and confirmed no customer data or credentials were exposed.

But this marks the company's second major data blunder in under a week. Days earlier, descriptions of an upcoming AI model called "Mythos" and other internal documents were discovered in a publicly accessible data cache (Source: Fortune).

For a company positioning itself as the safety-first AI lab and preparing for a potential IPO, shipping its own source code twice in one week presents a trust problem no press statement can fully patch.

Match Group and its subsidiary OkCupid settled with the Federal Trade Commission on Monday over allegations that the dating platform secretly shared nearly three million users' photos, location data, and personal details with facial recognition company Clarifai (Source: MediaPost).

The data transfer happened in 2014, after Clarifai's founder requested the dataset. The connection was personal: OkCupid's founders were financial investors in Clarifai. No contractual restrictions were placed on how the information could be used (Source: ClaimDepot).

The FTC alleged that OkCupid had represented to users that it only shared personal information in limited scenarios, such as with service providers or business partners. Clarifai was neither. The agency also accused OkCupid of "extensive efforts to obscure and deny the data sharing," including obstructing the FTC's Civil Investigative Demand (Source: ID Tech Wire).

The settlement carries no monetary fine. It permanently bars misrepresentation of data practices and requires 10 years of compliance reporting, but does not require OkCupid to notify affected users or offer any remedy to those whose data was shared (Source: The New Republic).