Good morning. It’s Thursday, November 27. Before the turkey hits the table, today we’re covering:
MIT study finds AI can already replace 11.7% of U.S. workforce
What to know about a recent Mixpanel security incident
Crypto Exchange Upbit Hit by Exploit After $10.3 Billion Deal
HP to cut up to 6,000 jobs by 2028 in AI-driven restructuring as tech layoffs continue
EU states back away from forcing Big Tech to detect and remove child pornography
Let’s dive in
MIT study finds AI can already replace 11.7% of U.S. workforce
By MacKenzie Sigalos via CNBC
MIT researchers, using the Iceberg Index labor-simulation tool, estimate today’s AI systems could already replace 11.7% of the U.S. workforce, exposing up to $1.2 trillion in wages across finance, health care and professional services, far beyond the visible tech layoffs that account for just 2.2% of wage exposure.
The Iceberg Index builds a “digital twin” of the labor market, modeling 151 million workers, 32,000 skills, 923 occupations and 3,000 counties to show where current AI can perform specific tasks, reveal exposure in routine roles like HR, logistics, finance and office administration, and let policymakers run “what-if” scenarios down to the zip code.
States including Tennessee, North Carolina and Utah are already using Iceberg’s county-level simulations to shape AI workforce action plans, identify exposure hotspots across all 50 states (including rural regions), and design targeted reskilling, training and infrastructure investments that aim to augment, rather than hollow out, key physical-work sectors.
𝕏: Huge wave is coming that will make every past technological shift look microscopic. “Massachusetts Institute of Technology on Wednesday released a study that found that artificial intelligence can already replace 11.7% of the U.S. labor market, or as much as $1.2 trillion in wages across finance, health care and professional services.” - Dr Singularity (@Dr_Singularity)
What to know about a recent Mixpanel security incident
OpenAI
Incident scope: A security breach at Mixpanel, OpenAI’s third-party web analytics provider for platform.openai.com, exposed limited analytics/profile data for some API users, but did not touch OpenAI’s own infrastructure.
What was not affected: No ChatGPT accounts, API data, prompts, or outputs, nor any passwords, API keys, payment info, IDs, session/auth tokens, or other sensitive credentials were exposed, so password resets or key rotation aren’t being recommended.
Response and next steps: OpenAI has removed Mixpanel, is reviewing the impacted datasets, notifying affected users and org admins by email, coordinating with Mixpanel, recommending multi-factor authentication, and committing to update this FAQ if new, material information emerges.
𝕏: Just got an email from OpenAI. Their analytics vendor Mixpanel got breached, names, emails, locations, and account metadata of API users exposed. Not chat logs this time, but it’s a reminder of what’s coming. People tell AI things they wouldn’t put in emails. Medical questions, relationship problems, business strategies, personal confessions… - Marouane Lamharzi Alaoui (@marouane53)
The best way to reach new readers is through word of mouth. If you click THIS LINK in your inbox, it’ll create an easy-to-send pre-written email you can just fire off to some friends.
Crypto Exchange Upbit Hit by Exploit After $10.3 Billion Deal
By Sidhartha Shukla via Bloomberg
Upbit, South Korea’s leading crypto exchange, suspended deposits and withdrawals after detecting an unauthorized transfer of roughly $30 million in Solana network-based assets to an external wallet on Nov. 27.
The stolen Solana-linked funds were first estimated at 54 billion won, later revised to about 44.5 billion won to reflect asset prices at the time of the exploit.
The incident comes just one day after Upbit unveiled a $10.3 billion takeover by Naver Corp, casting a shadow over the high-profile acquisition deal.
𝕏: UPBIT CONFIRMS ~$37M SOLANA HACK, SAYS ALL LOSSES WILL BE FULLY COVERED BY THE EXCHANGE, WITH NO IMPACT ON USER FUNDS. WILD - 0xMarioNawfal (@RoundtableSpace)
HP to cut up to 6,000 jobs by 2028 in AI-driven restructuring as tech layoffs continue
By Nickie Louise via Tech Startups
HP plans to cut 4,000–6,000 jobs globally by fiscal 2028 as part of an AI-driven restructuring aimed at $1 billion in cost savings over three years, on top of earlier layoffs of 1,000–2,000 employees, sending shares down about 5.5% in extended trading.
The move slots into a wider Big Tech retrenchment, with at least 237 tech companies cutting 114,000+ jobs in 2025, including Amazon (14,000 jobs), Google (hundreds across design, Android, Pixel, Chrome), and Microsoft (9,000 jobs), as firms redirect spending toward AI initiatives.
While over 30% of HP’s Q4 PC shipments are AI-enabled, rising DRAM and NAND memory costs are pressuring margins, prompting actions like qualifying lower-cost suppliers and trimming memory configurations; HP now guides fiscal 2026 EPS to $2.90–$3.20 (below street at $3.33) despite Q4 revenue of $14.64 billion slightly beating expectations.
𝕏: This isn’t AI replacing workers. HP is bleeding in every core business and using “AI” as cover for cuts to hit Wall Street targets. Consumer PC revenue dropped 7% this quarter. Printer hardware down 12%. Meanwhile memory costs jumped 15-18% and are eating their margins alive. The company’s stuck between a dying printing business and a saturated PC market where they missed the early AI hardware wave… - Aakash Gupta (@aakashg0)
EU states back away from forcing Big Tech to detect and remove child pornography
Reuters
EU member states agreed a common position on online child protection rules that stops short of forcing Big Tech to detect and remove child sexual abuse material, instead requiring platforms to assess risks, take preventive measures, and leaving penalties to national authorities.
The softer stance, seen as a win for Google, Meta, and privacy / anti-surveillance advocates, diverges from the European Parliament’s tougher 2023 position, allows voluntary content scanning beyond April when a privacy exemption expires, and creates an EU Centre on Child Sexual Abuse to support enforcement and victims.
Separately, the European Parliament called (non-binding) for EU-wide minimum social media ages amid rising adolescent mental health concerns, mirroring planned or proposed under-16 social media bans in Australia, Denmark, and Malaysia.
𝕏: In plain English: The Council acted like it was still the East German era, stripping 450 million EU citizens of their right to privacy. - Rob Roos (@Rob_Roos)
Trending in AI
China’s tech giants take AI model training offshore to tap Nvidia chips
Jeff Bezos’ New AI Venture Quietly Acquired an Agentic Computing Startup
Italy’s Competition Watchdog Broadens WhatsApp AI Policy Probe
Thanks for sticking with us to the very end of Newslit Daily and letting us be part of your morning.
On this Thanksgiving, I’m truly thankful to have you as a reader.
As always, if you have questions or feedback, just hit reply.
Enjoy the day, and see you tomorrow!
How was today’s email?
