The Slop Report - May 6, 2026
Your daily digest of AI-generated content news from around the web. All signal, no slop.
1. Doctors’ growing AI deepfakes problem
Axios - · May 6
Summary The American Medical Association is calling for new privacy and transparency laws after
AI-generated deepfake videos featuring real doctors without their consent are being used on social media to promote dubious products and spread medical misinformation. This trend threatens public trust in healthcare providers and could enable insurance fraud, data theft, and patient harm. The issue underscores the need for regulatory action to protect both physicians’ identities and the integrity of medical information online.
2. Approaching Half of New Podcasts Appear to Be AI Slop
Futurism - · May 5
Summary Approximately 39% of new podcasts created in the past nine days are AI-generated,
according to data from the Podcast Index, with companies like Inception Point mass-producing thousands of low-quality AI shows per week at minimal cost. These AI podcasts feature AI-written scripts and synthesized voices that often contain errors and vapid content, with shows like one called “Lawn” offering little useful information. Some podcast hosting platforms like RSS.com are fighting back by requiring minimum listener thresholds and pulling ads from detected “slop,” though it remains unclear how many people actually listen to these AI-generated shows or how much revenue they generate.
3. Google Is Testing New Bot Authorization Standard via @sejournal, @martinibuster
Search Engine Journal - · May 5
Google is testing Web Bot Auth, an experimental protocol that uses cryptographic verification to help websites confirm that automated traffic is genuinely from the bot or service claiming to send it, rather than relying on easily-spoofed identifiers like user-agent strings. The protocol works by having bots include digital credentials in their requests that websites can cryptographically verify against standardized key files, making it harder for malicious bots to impersonate legitimate services. This matters because it could give site owners a more reliable way to distinguish legitimate crawlers and automated services from imposter bots while reducing the burden of manual security setup between websites and bot operators.
4. Axios interview: Scale AI CEO Jason Droege pushes “reliablity” reality
Axios - · May 6
Summary Scale AI CEO Jason Droege told Axios that current AI systems are too unreliable for
mission-critical applications in business, military, and government contexts where errors carry severe consequences. Droege took over the role last June after founder Alexandr Wang became Meta’s first chief AI officer, with Meta acquiring a 49% stake in the company. This statement matters because it highlights a significant limitation of AI adoption in high-stakes environments and reflects the priorities of a major AI infrastructure company as the technology matures.
5. ChatGPT for Excel and Google Sheets is now out of beta, and it’s kind of a big deal
Digital Trends - · May 6
Summary OpenAI has officially launched ChatGPT integration for Microsoft Excel and Google Sheets
out of beta, making it available globally to all users across both platforms. The AI feature allows users to build spreadsheets from scratch using text prompts, clean up messy data, create formulas, and ask questions about existing spreadsheets without leaving the application. This matters because it significantly lowers the barrier to entry for spreadsheet work, helping both casual and advanced users save time on complex data tasks and formula creation.
6. ‘I thought he was going to hit me,’ OpenAI co-founder says of Musk
BBC Technology - · May 6
OpenAI president Greg Brockman testified in court that during a heated 2017 meeting, Elon Musk became aggressive when Brockman rejected his proposal for greater control of the company, prompting Musk to threaten to withdraw funding. Brockman’s testimony is part of Musk’s ongoing lawsuit against OpenAI attempting to reverse its transition to a for-profit structure, with Musk arguing he was unaware of these plans despite evidence suggesting otherwise. The case reflects a bitter dispute stemming from Musk’s departure from OpenAI, which he co-founded but left before it became a leading AI company through ChatGPT’s success.
7. Microsoft Copilot Cowork goes mobile and it’s ready to take tasks off your plate
Digital Trends - · May 6
Microsoft announced that Copilot Cowork, its AI assistant designed to autonomously complete tasks rather than just answer questions, is now available on iOS and Android devices. The update includes new features like customizable “skills” that let users standardize workflows and deeper integrations with Microsoft 365 and third-party tools like Miro and monday.com. This matters because it allows users to delegate work from their phones and have AI continue progressing on tasks across devices, representing a shift toward AI agents that actively execute work rather than simply provide information.
8. CO2 Levels In the Atmosphere Hit ‘Depressing’ New Record
Slashdot - · May 6
Summary Atmospheric CO2 levels hit a record 431 parts per million in April 2026, measured at
NOAA’s Mauna Loa Observatory—up from under 320 ppm since measurements began in 1958. Climate scientists call the milestone “depressing” though expected, noting that while U.S. emissions briefly declined in 2023-2024, they reversed in 2025 partly due to increased electricity demand from AI data centers. The story matters because it demonstrates that despite some renewable energy growth, global carbon emissions continue accelerating at an alarming rate.
9. Your ChatGPT history is a personality test you didn’t know you were taking
Digital Trends - · May 6
Summary Researchers at ETH Zurich trained an AI model that can predict personality traits from
ChatGPT conversation logs with concerning accuracy, analyzing 62,090 real conversations from 668 users across five personality dimensions. The study found that certain topics (mental health discussions, religious conversations, mood-related chats) made personality predictions particularly easy, and users who interact more frequently with ChatGPT become easier to profile. This matters because with over 800 million monthly ChatGPT users and OpenAI’s recent integration of ads, companies could use personality profiles derived from chat history for targeted advertising, personalized manipulation, or influence campaigns.
10. Brockman Rebuts Musk’s Take On Startup’s History, Recounts Secret Work For Tesla
Slashdot - · May 6
Summary OpenAI President Greg Brockman testified in federal court, rebutting Elon Musk’s account
of the startup’s early years by stating that Musk never secured commitments about corporate structure and that OpenAI remains nonprofit-governed. Brockman also revealed that Musk had enlisted OpenAI employees to perform months of unpaid work on Tesla’s Autopilot self-driving technology in 2017, and testified that Musk became angry during 2017 negotiations over a for-profit arm, even tearing a Tesla painting off the wall. The testimony is part of an ongoing trial that will continue with testimony from Shivon Zilis, a former OpenAI board member and mother of four of Musk’s children.
11. Elon Musk Wanted OpenAI to Go Commercial, Greg Brockman Testifies
NY Times Tech - · May 6
Greg Brockman, OpenAI’s president, testified in court that Elon Musk—who co-founded the company—wanted to alter its nonprofit structure during his time with the organization. This testimony is part of Musk’s legal case against OpenAI, likely regarding disputes over the company’s governance or direction after it transitioned toward for-profit operations. The case matters because it addresses questions about OpenAI’s founding principles, Musk’s role in its early development, and potential conflicts over how the company evolved into one of the most powerful AI organizations in the world.
12. OpenAI Launches Self-Serve Ads Manager for ChatGPT via @sejournal, @brookeosmundson
Search Engine Journal - · May 6
OpenAI has launched a beta self-serve Ads Manager for ChatGPT, allowing U.S. advertisers to directly create and manage campaigns with new cost-per-click bidding and conversion measurement tools, moving beyond the previous managed partnership model. The update includes features like budget control, campaign performance monitoring, and conversion tracking through APIs and pixels, while maintaining privacy protections that prevent advertisers from accessing user conversations. This launch makes ChatGPT advertising more accessible to smaller businesses and performance marketers, establishing it as a legitimate advertising channel comparable to traditional media buying platforms.
13. Apple to pay up to $95 to some US iPhone buyers over AI lawsuit
BBC Technology - · May 6
Apple will pay up to $250 million to settle a class action lawsuit from iPhone 15 and iPhone 16 buyers who claim the company falsely advertised AI features called Apple Intelligence that weren’t actually available at the time of purchase. The settlement, which does not require Apple to admit wrongdoing, will distribute between $25-$95 to affected US customers, resolving claims that Apple marketed non-existent capabilities like an enhanced Siri assistant to compete in the AI arms race with companies like OpenAI. This matters because it highlights regulatory and consumer pushback against tech companies making unfulfilled AI promises during the industry’s rapid expansion.
14. Apple agrees to pay $250m after falsely claiming AI-powered Siri was ‘available now’
The Guardian Tech - · May 5
Apple agreed to pay $250 million to settle a class-action lawsuit for misleading consumers about AI- powered Siri capabilities that were advertised as “available now” in late 2024 but have yet to be fully released. The settlement covers approximately 36 million eligible iPhone 16, 15 Pro, and 15 Pro Max devices purchased in the US, with each claimant potentially receiving $25-$95 per device. This matters because it highlights how major tech companies may face legal consequences for overpromising AI features to boost sales, setting a precedent as AI becomes increasingly central to consumer purchasing decisions.
15. OpenAI president forced to read his personal diary entries to jury
Ars Technica - · May 5
In an ongoing trial where Elon Musk alleges OpenAI abandoned its nonprofit mission to enrich executives, OpenAI president Greg Brockman has been forced to publicly read deeply personal journal entries in court, including passages where he discussed making money and earning $1 billion. Musk’s legal team is using the 100-page journal—originally intended only for Brockman’s private reflection and containing stream-of-consciousness musings—to portray him as a greedy executive focused on personal wealth rather than OpenAI’s charitable mission. The case highlights the clash between Musk’s vision for OpenAI and the company’s current leadership, with Brockman’s testimony central to determining whether the organization genuinely prioritizes its stated mission.
16. Google Home’s Gemini AI can handle more complicated requests
The Verge AI - · May 5
Google has upgraded Gemini for Home to version 3.1, enabling the smart home assistant to handle more complex, multi-step tasks and combine multiple requests in a single voice command. The update improves the AI’s ability to interpret natural language, manage calendar events, and identify devices correctly, addressing previous bugs in the system. Google is also rolling out new features including Ask Home on Web for remote smart home management and improved notification controls with quick-action buttons.
17. Silicon Valley bets $200M on AI data centers floating in the ocean
Ars Technica - · May 5
Silicon Valley investors, including Palantir co-founder Peter Thiel, are backing Panthalassa with $200 million to build AI data centers on floating ocean nodes powered by wave energy, as tech companies struggle with land-based data center development challenges. The wave-powered spherical nodes would directly compute AI workloads and transmit results via satellite, while gaining cooling advantages from ocean water. However, significant hurdles remain, including satellite bandwidth limitations, coordination challenges between distributed nodes, and maintenance complications in remote ocean locations.
18. Anthropic reportedly agrees to pay Google $200 billion for chips and cloud access
Engadget - · May 5
Summary Anthropic has agreed to pay Google $200 billion over five years for cloud computing
resources and chips needed to develop its Claude AI models, according to The Information. This deal is part of a larger trend where major tech companies like Google, Amazon, Microsoft, and Oracle have secured roughly $2 trillion in commitments from AI startups like Anthropic and OpenAI, betting that these companies’ massive infrastructure needs would become profitable long-term investments. While these arrangements have proven lucrative for cloud and chip providers so far, experts question their sustainability given rising data center demands and hardware constraints.
19. Xbox CEO scraps Copilot AI for consoles to refocus the platform on gameplay-first experiences
Digital Trends - · May 5
Xbox CEO Asha Sharma announced that Microsoft is canceling Copilot AI for Xbox consoles and mobile, ending development of the AI assistant feature that was designed to provide in-game tips and guidance. The decision reflects Xbox’s new strategy under Sharma to prioritize gameplay-first experiences and reduce friction for players and developers, moving away from experimental features that don’t directly improve player experience. This cancellation is part of a broader Xbox reset that includes dropping the “Microsoft Gaming” branding and refocusing on core gaming priorities rather than flashy AI tools.
20. Scott Turow’s latest real-life legal thriller: Suing Meta for copyright infringement
NPR Technology - · May 5
Five major publishers (Hachette, Macmillan, McGraw Hill, Elsevier, and Cengage) and bestselling author Scott Turow filed a class-action lawsuit against Meta and CEO Mark Zuckerberg, alleging the company illegally trained its Llama AI models on millions of copyrighted books and journal articles sourced from pirate websites like LibGen and Anna’s Archive. The plaintiffs claim Meta deliberately bypassed licensing agreements to gain a competitive advantage in AI development and are seeking statutory damages, an injunction, and destruction of infringing materials. This lawsuit represents a major legal challenge to how tech companies source training data for generative AI and could have significant implications for copyright protection in AI development.
21. Xbox is ditching Microsoft’s Copilot AI
Engadget - · May 5
Microsoft is removing Copilot AI from Xbox consoles and its mobile app, with CEO Asha Sharma citing the need to move faster and better serve players and developers. The decision reverses previous plans to expand the in-game assistant across Xbox platforms and comes as Microsoft has been facing criticism over Copilot’s implementation across its products. The move signals that Xbox’s new leadership under Sharma will prioritize direct community connection over AI features, though AI may still play a behind-the-scenes role in developer tools and operations.
22. Google DeepMind Workers Vote To Unionize Over Military AI Deals
Slashdot - · May 5
Google DeepMind employees in London have voted to unionize, demanding the company recognize the Communication Workers Union and Unite the Union as their representatives, primarily to block the AI lab from providing technology to US and Israeli militaries. The unionization effort was sparked by Alphabet’s February 2025 removal of ethical AI guidelines that previously barred weapons development and surveillance uses. If Google refuses to negotiate, workers plan to seek arbitration and may demand the company withdraw from its Israeli military contract and provide greater transparency on AI product use.
23. Apple Reaches $250 Million Settlement Over Claims It Misled People on A.I.
NY Times Tech - · May 5
Summary iPhone owners are eligible for compensation of $25 to $95 through a settlement over
Apple’s marketing claims about Apple Intelligence, its AI system. The settlement addresses allegations that Apple overstated the capabilities and availability of its artificial intelligence features. This matters because it reflects growing scrutiny of tech companies’ AI marketing practices and establishes precedent for holding firms accountable when AI capabilities don’t match promotional claims.
24. Apple plans to make iOS 27 a Choose Your Own Adventure of AI models
TechCrunch AI - · May 5
Apple plans to introduce “Extensions” in iOS 27, allowing iPhone users to choose which third-party AI models power on-device features like Siri and Writing Tools, with Google and Anthropic models currently being tested alongside the existing ChatGPT integration. This move positions Apple as offering users choice in their AI experience rather than competing directly with rivals by building proprietary AI infrastructure, aligning with incoming CEO John Ternus’s strategy to leverage Apple’s existing hardware as an AI-centric platform.
25. US to safety test new AI models from Google, Microsoft, xAI
BBC Technology - · May 5
The US Department of Commerce announced that Google, Microsoft, and xAI have agreed to voluntarily submit their AI models for safety testing through the Center for AI Standards and Innovation (CAISI) before public release—expanding earlier agreements with OpenAI and Anthropic. The evaluations will assess the models’ capabilities, security, and potential risks related to national security and public safety. This represents a notable shift toward government oversight under the Trump administration, which has generally favored a hands-off approach to AI regulation.
25 stories sourced from Ars Technica, Axios, BBC Technology, Digital Trends, Engadget, Futurism, NPR Technology, NY Times Tech, Search Engine Journal, Slashdot, TechCrunch AI, The Guardian Tech, The Verge AI. The Slop Report is published daily. Subscribe via RSS.