Weekly Slop Roundup — May 16, 2026
The week’s biggest stories on AI-generated content, deepfakes, synthetic media, and information quality.
50 stories from 18 sources this week.
1. The Download: deepfake porn’s stolen bodies and AI sharing private numbers
MIT Technology Review
Summary MIT Technology Review’s newsletter covers three major AI stories: deepfake pornography is
being created using the bodies of adult content creators whose work is used without consent to train AI systems, leaving them with minimal legal protections; generative AI chatbots like Google’s Gemini are inadvertently exposing people’s private phone numbers by pulling personally identifiable info
2. AI Slopification and Writing
Hacker News
I’d be happy to help summarize an AI news story for you, but I don’t see any article text, link, or title in your message. Could you please share the article, link, or headline you’d like me to summarize?
3. Netflix has its own AI studio now, and AI-generated content is coming for your feed whether you like it or not
Digital Trends
Netflix has launched a new internal AI studio called INKubator, led by former DreamWorks executive Serrena Iyer, to produce AI-generated animated shorts and specials. The studio, which quietly began operations in March 2026, represents Netflix’s expansion beyond using AI for recommendations into actually creating content, with job listings indicating plans for scalable multi-show environments and
4. Federal judge holds back on Anthropic’s $1.5bn author settlement
The Next Web
Summary A federal judge in San Francisco declined to approve Anthropic’s proposed $1.5 billion
settlement with authors who claim the company illegally used pirated books to train its Claude AI models, requesting more details on lawyer fees and lead-plaintiff payments before granting final approval. The settlement, which would be the largest copyright settlement in U.S. history, covers approxima
5. AI writing hits a ceiling
Axios
Summary According to a new analysis by digital marketing agency Graphite, the percentage of AI-
generated content on the web has plateaued at around 50% and has remained relatively stable for over a year, contrary to earlier fears that AI-generated writing would overwhelm the internet following ChatGPT’s release. The study, which sampled URLs from Common Crawl using three AI-detector tools, sug
6. OpenAI is bringing in the mighty Codex tool to the ChatGPT app on your phone
Digital Trends
Summary OpenAI has launched Codex, its AI-powered coding agent, in the ChatGPT mobile app for iOS
and Android, allowing developers to remotely control and monitor coding sessions running on their computers from their phones. The mobile version functions as a remote control that lets users review threads, approve commands, check terminal output, and inspect code changes without being at their de
7. Jury deliberations to begin in OpenAI nonprofit trial after Musk skips closing for Beijing
The Next Web
Elon Musk is suing OpenAI founders Sam Altman and Greg Brockman, claiming they breached charitable trust by converting OpenAI from a nonprofit to a for-profit entity worth $350 billion without his consent, seeking $134 billion in disgorgement and removal of the executives. Jury deliberations began Monday in Oakland federal court after three weeks of testimony, with Musk absent from closing argumen
8. xAI introduces its coding agent called Grok Build
Engadget
xAI has launched Grok Build, a coding agent designed to compete with rivals like Anthropic’s Claude Code, initially available only to SuperGrok Heavy subscribers ($300/month) in beta. The move represents xAI’s effort to catch up in AI coding capabilities after CEO Elon Musk acknowledged the company had fallen behind competitors, though Grok has faced reputational challenges including generating no
9. Musk Accused of ‘Selective Amnesia’, Altman of Lying As OpenAI Trial Nears End
Slashdot
Elon Musk is suing OpenAI and its CEO Sam Altman, claiming they transformed the nonprofit into a for-profit vehicle to enrich themselves rather than fulfilling their mission to build safe AI for humanity’s benefit. During closing arguments in the Oakland trial, Musk’s lawyers attacked Altman’s credibility and accused OpenAI and Microsoft of wrongdoing, while OpenAI’s defense argued Musk waited too
10. OpenAI Trial Heads to Jury After Closing Arguments in Musk vs. Altman Case
NY Times Tech
Elon Musk is suing OpenAI, alleging the company breached its founding agreement by prioritizing profit over its nonprofit mission after transitioning to a for-profit structure and partnering with Microsoft. The case went to federal trial with closing arguments from both sides on Thursday, and a jury of nine will begin deliberations next week to decide the outcome. The lawsuit matters because it co
11. OpenAI Considers Legal Action Against Apple in Strained Relationship
NY Times Tech
I don’t have access to the full article content or URL you’re referring to. To provide an accurate summary with specific details about OpenAI’s complaints regarding Apple’s ChatGPT integration and the legal dispute with Elon Musk, could you share the article link or provide more of the text? This will help me give you a precise, factual summary rather than speculation based only on the headline.
12. High-stakes courtroom drama of Musk v OpenAI hears closing arguments
The Guardian Tech
Summary Elon Musk’s lawsuit against OpenAI and CEO Sam Altman reached closing arguments in an
Oakland federal court on Thursday, with a nine-person jury now set to decide whether the AI firm breached founding agreements and unjustly enriched itself by restructuring from non-profit to for- profit status. Musk claims Altman and OpenAI violated their founding agreement and bilked him out of money,
13. AI could put people off tech jobs and hurt the economy, warns Raspberry Pi boss
BBC Technology
Summary Eben Upton, founder of Raspberry Pi, warned that overestimating AI’s capabilities could
discourage young people from pursuing tech careers, potentially worsening skill shortages and damaging economic growth. He cautioned against hype around AI tools like ChatGPT replacing computing jobs, arguing that such claims lack data and could undo efforts to encourage people into technology fields
14. UK Antitrust Regulator Is Officially Investigating Microsoft Office
Slashdot
The UK’s Competition and Markets Authority has launched a formal investigation into Microsoft’s practice of bundling Windows, Office, Teams, Copilot, and other products together, examining whether this harms competition and consumer choice in the business software market. The regulator will also review Microsoft’s cloud licensing practices, with a conclusion expected by February, potentially leadi
15. Show HN: Parse LLM Markdown streams incrementally on the server or client
Hacker News
This is a GitHub repository announcement for a new TypeScript markdown parser built by Nimeshan Nayaju that specializes in streaming/incremental parsing—useful for processing LLM-generated markdown in real-time. The parser converts markdown into a fully-typed node tree following CommonMark specifications and supports GitHub Flavored Markdown (GFM) tables, allowing developers to emit finalized cont
16. Edge browser on mobile gets a huge upgrade that makes it a worthy pick over Chrome
Digital Trends
Microsoft has released a major Copilot update for Edge mobile that allows users to compare information across multiple open tabs and organize browsing history into topic-based “Journeys,” positioning it as a competitor to Chrome ahead of Google’s Gemini-powered Chrome update expected in June. The new features—including cross-tab reasoning, voice/vision capabilities, and a redesigned home screen—gi
17. Figma’s numbers say AI is a tailwind. Its stock price says the market isn’t sure.
The Next Web
Figma reported Q1 2026 revenue of $333.4 million (46% year-on-year growth), beating expectations and raising full-year guidance by $55 million, with the stock jumping 8% after hours. The critical success metric was that 75% of higher-tier users continued paying for AI credits after the company began enforcing limits on March 18, though 5% of those users churned entirely. This demonstrates early mo
18. Behold, the Elon Musk jackass trophy
The Verge AI
Summary During the ongoing lawsuit between Elon Musk and Sam Altman over OpenAI’s future, a
humorous “jackass trophy” surfaced in court that OpenAI employees had given to researcher Josh Achiam—allegedly commemorating an incident where Musk called Achiam a jackass for questioning whether racing ahead of Google on AI development was wise. The trophy became relevant because it undermines Musk’s c
19. OpenAI says Codex is coming to your phone
TechCrunch AI
OpenAI has integrated its Codex coding tool into the ChatGPT mobile app, allowing developers to monitor and manage their development workflows remotely from iOS and Android devices. The update, now available in preview across all plans, enables users to view live Codex environments, review outputs, approve commands, and manage tasks from their phones. This move reflects intensifying competition be
20. Elon Musk Flees OpenAI Trial as Tide Turns Against Him
Futurism
Elon Musk left the country to visit China with President Trump while under court order to remain available for testimony in his ongoing lawsuit against OpenAI and Sam Altman, missing closing arguments despite a judge’s explicit directive to stay on “recall status.” Musk’s absence and his combative courtroom performance—during which he contradicted previous statements, admitted to not reading key d
21. How Trump may be changing his stance on AI regulation
NPR Technology
Summary The Trump administration, which initially prioritized AI innovation over regulation and
dismantled Biden-era AI safety measures, has recently shifted its rhetoric to emphasize the need for careful oversight and safety considerations. The shift was triggered by AI company Anthropic’s warning about their powerful Mythos model posing cybersecurity risks, prompting White House officials inc
22. Cerebras, A.I. Chip Maker, Rises 89% in Market Debut as Tech IPOs Ramp Up
NY Times Tech
Cerebras IPO Cerebras, a Silicon Valley AI chip manufacturer, went public on Thursday, joining
other major AI companies like SpaceX, OpenAI, and Anthropic in pursuing public market listings. This move reflects growing investor confidence in the AI hardware sector and signals a potential shift toward broader commercialization of AI infrastructure as these companies seek capital for expansion and
23. Anthropic tosses agents into the API billing pool
The Register
Anthropic is separating Claude subscription usage into two pools starting June 15, 2026: interactive use (human-in-the-loop) remains on the current subscription rate, while programmatic use (agents, APIs, headless mode) gets its own dedicated budget funded by the subscription fee but billed at costlier API rates. This move, framed as responsive customer service, represents the company’s continued
24. OpenAI’s Codex is now in the ChatGPT mobile app
The Verge AI
Summary OpenAI has integrated its Codex desktop AI tool into the ChatGPT mobile app, allowing
users to control and monitor code-writing tasks on their computer from their iOS or Android phone in real time. The feature, rolling out now as a preview across all ChatGPT plans, enables users to approve commands, review outputs, and receive live updates including screenshots and test results. This mo
25. Claude Helps Recover Locked $400K Bitcoin Wallet After 11 Years
Slashdot
A Bitcoin user recovered 5 BTC (worth ~$400,000) that had been inaccessible for 11 years with help from Anthropic’s Claude AI. After finding an old mnemonic seed phrase, the user uploaded their entire college computer files to Claude, which discovered an older wallet backup from 2019 and identified a bug in the password recovery tool’s configuration that had prevented previous recovery attempts. T
26. 4 questions you should ask to catch programmatic ad fraud
MarTech
Summary The article discusses how programmatic advertising fraud is costing the industry $26.8
billion annually, with connected TV (CTV) advertising particularly vulnerable to sophisticated schemes like device and app spoofing. Sean Nowlin, CEO of SpotlightIQ, argues that marketers’ biggest risk isn’t the fraud itself but their passive assumption that intermediaries (DSPs, agencies, publishers)
27. Podcast: The Chinese Deepfake Software Powering Scams
404 Media
Summary 404 Media obtained Haotian AI, Chinese-language deepfake software that enables real-time
video manipulation during video calls on platforms like Microsoft Teams, WhatsApp, and Zoom, allowing users to impersonate others during scams. The podcast episode also covers a man discovering $1 million worth of Yu-Gi-Oh cards and discusses how the AI-driven hard drive shortage is hampering intern
28. The shock of seeing your body used in deepfake porn
MIT Technology Review
Summary Jennifer, a former adult content creator, discovered her body had been used without
consent in a deepfake video with another woman’s face, highlighting a largely overlooked harm in the deepfake pornography crisis. The article explores how generative AI and “nudify” apps increasingly use adult performers’ bodies as training data to create non-consensual synthetic content, threatening bot
29. Arborium is AI slopware and should not be trusted
Hacker News
Evie On-line criticizes Arborium, a new tree-sitter-based syntax highlighting tool created by respected developer Amos Wenger, for being poorly documented and unreliable. After attempting to integrate Arborium into her website, Evie encountered multiple technical failures including incompatibility with Deno runtime, undocumented configuration options, and hacky dynamic code importing that didn’t f
30. Show HN: Learn algebra together, no AI slop, no ads or freemium, no registration
Hacker News
I don’t have access to the specific article you’re referring to. To provide an accurate summary of what happened, who’s involved, and why it matters, I would need either a link to the article, the full text, or more details about where this story was published. If you can share the article link or text, I’d be happy to summarize it for you.
31. Ask HN: What are you working on (non-AI)?
Hacker News
I appreciate you sharing this, but this appears to be a community post or forum discussion rather than a news story about AI. The text is someone’s opinion about moderating content in a “What are you working on?” community—they’re expressing concern that AI-generated or AI-heavy projects are dominating discussion posts and requesting more focus on human-created work. This is a meta- discussion abo
32. Pope Leo sets Catholics on collision course with AI
Axios
I cannot provide an accurate summary of this article because Pope Leo XIV does not exist—the current Pope is Francis, and there has been no Pope Leo XIV in Catholic Church history. The article appears to be fictional or speculative. If you have a real AI news story you’d like summarized, I’d be happy to help.
33. The SpaceX IPO is already upending the stock market
Axios
I can see this article appears to be behind a paywall or incomplete. Based on the available excerpt, here’s what I can tell you: SpaceX, Elon Musk’s privately-held company currently valued at over $1 trillion, is expected to go public next month with a potential valuation reaching $2 trillion, marking what could be a significant shift in public markets dominated by AI companies. The IPO is part o
34. The energy squeeze behind the Iran war and AI boom
Axios
I don’t have access to the full article, but based on the excerpt provided: Energy has emerged as the world’s most critical constraint, with oil supply disruptions from Middle East tensions driving inflation and geopolitical instability, while the AI industry’s explosive growth is creating unprecedented electricity demand that existing power grids cannot support. This matters because energy availa
35. Microsoft brings tab intelligence to Edge browser, and I dearly wish Apple would add it to Safari
Digital Trends
Summary Microsoft has added “tab intelligence” to its Edge browser, allowing Copilot to read and
synthesize information across all open tabs to answer user questions without switching between them—for example, comparing travel options across multiple research tabs. The feature can also reference browsing history and past chats for more contextual answers. The author argues this capability would
36. Cerebras raises $5.55bn in the biggest US tech IPO since Snowflake
The Next Web
Summary Cerebras Systems, a chip company specializing in large-scale AI inference processors,
completed the largest U.S. tech IPO since 2020 by raising $5.55 billion at a $56.4 billion valuation, with shares pricing at $185 and beginning to trade Thursday on Nasdaq. The company’s IPO momentum was driven by a landmark multi-year contract with OpenAI worth over $20 billion at full expansion, whic
37. This memory chip works at 700 degrees Celsius. The startup behind it is already building AI chips that compute where GPUs cannot.
The Next Web
USC researchers have developed a memristor memory chip that operates reliably at 700 degrees Celsius, far exceeding previous temperature limits and opening possibilities for extreme-environment applications like Venus exploration. TetraMem, the startup commercializing the technology, has already moved its room-temperature AI inference chips to production-scale 300mm wafers with support from SK hyn
38. OpenAI says no user data was touched in the TanStack npm worm
The Next Web
Summary On May 11, attackers compromised 42 TanStack npm packages (downloaded over 518 million
times collectively) by hijacking GitHub Actions runners mid-build and stealing OIDC tokens, enabling them to publish malicious code using legitimate release credentials—the first documented “npm worm” with a valid signature. OpenAI confirmed two employee devices were affected and some credential mater
39. Microsoft is quietly shopping for an OpenAI replacement
The Next Web
Microsoft is actively seeking acquisitions or partnerships with AI startups to reduce its dependence on OpenAI, following a contract renegotiation in April that ended their exclusive relationship. The company previously attempted to acquire code-generation startup Cursor but withdrew due to regulatory concerns, and is now in talks with Stanford-based Inception, which develops alternative diffusion
40. AI models are getting better at replacing cybersecurity pros on certain tasks
The Register
Summary The UK AI Security Institute found that advanced AI models like Claude Sonnet 4.5 and
GPT-5.5 are rapidly improving at performing cybersecurity tasks, with the time required to complete human-level work doubling approximately every 4-4.7 months (down from an 8-month estimate in late 2025). Claude Mythos Preview has demonstrated particularly strong performance, solving complex simulated
41. Clio’s $500M milestone arrives just as Anthropic ups the ante
TechCrunch AI
Clio, a Canadian legal tech company, announced its annual recurring revenue (ARR) reached $500 million, driven by AI integration that has accelerated growth since 2023. The milestone comes as competitor legal AI startups like Harvey ($190M ARR) and Legora ($100M ARR) also experience explosive growth, and as Anthropic—whose Claude model powers many of these platforms—launches its own competing lega
42. Microsoft is retiring Copilot Mode on Edge, because everything is Copilot Mode now
Engadget
Microsoft is retiring the dedicated Copilot Mode feature on Edge because its AI capabilities are now integrated throughout the browser itself on both desktop and mobile. The company has expanded Copilot’s functionality to include new features like Journeys (for saving projects), Vision and Voice capabilities, Study and Learn mode, Writing Assistant, and the ability to convert open tabs into podcas
43. Arena AI Model ELO History
Hacker News
I don’t see a complete news article here—this appears to be documentation for an AI model performance tracking tool rather than a news story. The text explains that Arena AI created a chart monitoring ELO ratings of flagship AI models over time to detect hidden performance degradation (“nerfs”) caused by updates, censorship, quantization, or safety filters that labs may implement post-launch. The
44. Fact-checkers ask federal judge to block Trump policy they say chills free speech
Poynter
Summary Fact-checkers, journalists, and researchers filed a federal lawsuit challenging a 2025
Trump administration visa policy that restricts entry to individuals deemed “complicit in censoring Americans,” arguing it violates free speech and due process rights. The policy has targeted five people, including the founder of the Center for Countering Digital Hate and co-founders of fact- checking
45. Microsoft’s Edge Copilot update uses AI to pull information from across your tabs
The Verge AI
Microsoft is rolling out a major update to Edge’s Copilot AI assistant that can now access information across all of a user’s open browser tabs to answer questions, compare products, summarize articles, and generate study materials or podcasts. The update also introduces new features like “long-term memory” for personalized responses, an AI writing assistant, screen-sharing on mobile, and a redesi
46. SOLAI Launches $399 Solode Neo Linux AI Computer
Slashdot
SOLAI has launched the Solode Neo, a $399 Linux-based mini PC featuring an Intel N150 processor designed for always-on AI agents, browser automation, and persistent developer workflows with local control and privacy features. However, the device’s low-power hardware is better suited for lightweight automation and cloud-assisted tasks rather than running large AI models locally, making it more of a
47. Software Developers Say AI Is Rotting Their Brains
Slashdot
Summary Software developers on platforms like Reddit and Hacker News are reporting that using AI
code generation tools is making them less skilled, not more productive. Developers cite frustrations including spending more time fixing buggy AI output than writing code themselves, losing the ability to solve problems independently, and accumulating technical debt through poorly vetted AI-generate
48. I’m not sold on Googlebook’s future, but it sure has two big wins I can’t ignore
Digital Trends
Google introduced Googlebook, a new category of AI-centric laptops centered around its Gemini intelligence system, Android integration, and ChromeOS, marking a shift from operating system- focused devices to “intelligence systems.” While the author remains skeptical about Googlebook’s long-term viability compared to established platforms like Apple and Windows, two features stand out: Magic Pointe
49. Anthropic butts in to small business, promises help with payroll and other core tasks
The Register
Anthropic launched Claude for Small Business, a plugin offering small business owners easy access to AI-powered tools for automating payroll, payments, and marketing through integrations with platforms like QuickBooks, PayPal, and HubSpot. However, users on Pro and Max plans should be aware that Anthropic may use their business data to train its AI models, unlike enterprise customers whose data is
50. Anthropic’s Cat Wu says that, in the future, AI will anticipate your needs before you know what they are
TechCrunch AI
Summary Anthropic’s Cat Wu, head of product for Claude Code and Cowork, discussed the company’s
product strategy at a recent conference, emphasizing that Anthropic focuses on staying at the AI frontier rather than competing with rivals like OpenAI, and expects to continue releasing multiple AI models annually. Wu also outlined her vision for the future where AI agents will anticipate user needs
Weekly roundup of 50 stories from 404 Media, Axios, BBC Technology, Digital Trends, Engadget, Futurism, Hacker News, MIT Technology Review, MarTech, NPR Technology, NY Times Tech, Poynter, Slashdot, TechCrunch AI, The Guardian Tech, The Next Web, The Register, The Verge AI. The Slop Report is published daily. Subscribe via RSS.