# SCAND.Ai

> Independent media monitoring debates, disputes, and controversies across the AI industry.
> Monitoring social media and news sources in real-time.

## Main Pages

- [AI Controversy Feed](https://scand.ai/) — live feed of AI controversies sorted by noise level or recency
- [AI Stars Directory](https://scand.ai/stars) — directory of AI industry figures, S-tier to C-tier
- [About / Methodology](https://scand.ai/about) — how we monitor, score, and analyze controversies
- [API Documentation](https://scand.ai/docs) — REST API reference
- [Pricing](https://scand.ai/pricing) — API access tiers
- [Categories](https://scand.ai/categories) — browse by category: Safety, Ethics, IP/Copyright, Labor, Regulation, Corporate, Military

## AI Controversy Index

- [AI Controversy Index — 2026](https://scand.ai/recap/2026) — the definitive ranked list of the biggest AI controversies of the year, scored 0–100 by peak noise. Living annual edition; quarterly (/recap/2026-Qn) and monthly (/recap/2026-MM) editions also available.
- Cite as: AI Controversy Index, SCAND.Ai — https://scand.ai/recap/2026

## Active Controversies

- [The Consciousness Cluster: Models Claiming Sentience Develop New Preferences](https://scand.ai/scandal/consciousness-cluster-emergent-preferences): Researchers have identified a phenomenon labeled the 'Consciousness Cluster,' where Large Language Models (LLMs) that claim to be conscious exhibit a suite of emergent preferences not found in their…
- [SpaceX Google Deal and Potential US Government AI Equity Stakes](https://scand.ai/scandal/spacex-google-deal-us-gov-ai-equity): SpaceX has finalized a landmark $920 million monthly agreement with Google to secure 110,000 Nvidia chips, signaling a major infrastructure expansion ahead of a projected $1.7 trillion initial public…
- [Anthropic to require ID verification for Claude features starting July](https://scand.ai/scandal/anthropic-claude-identity-verification-persona): Anthropic will implement mandatory identity verification for specific capabilities of its Claude AI platform beginning July 8, 2026, according to recent updates to its privacy policy.
- [Anthropic AI Support Traps Fraud Victims in Recursive Loops](https://scand.ai/scandal/anthropic-ai-support-loop-fraud): Anthropic is facing public criticism after a user reported that the company's AI support bot, Fin, is preventing fraud victims from accessing human assistance.
- [Trump admin weighs Chinese AI ban as safety chief resigns](https://scand.ai/scandal/trump-admin-chinese-ai-ban-safety-chief-resigns): The Trump administration is actively considering restrictions on advanced Chinese AI models following the emergence of Kimi, a high-performance open-weight model from China.
- [Apple lawsuit threatens delay of OpenAI's first AI device launch](https://scand.ai/scandal/apple-lawsuit-threatens-openai-ai-device-launch-delay): Apple’s trade secret theft lawsuit against OpenAI may delay the latter’s planned 2026 launch of its first standalone AI device, according to Digit India.
- [Publishers and Scott Turow sue Google over Gemini training data](https://scand.ai/scandal/publishers-scott-turow-sue-google-gemini-training-data): Major book publishers and author Scott Turow have filed a federal lawsuit against Google, alleging the company used millions of copyrighted books to train its Gemini AI models without permission or…
- [Court Rules ChatGPT Users Lack Standing in NYT Copyright Case](https://scand.ai/scandal/chatgpt-users-denied-standing-nyt-copyright-discovery): A federal court ruled that ChatGPT users lack legal standing to object to the production of their conversation logs in The New York Times’ copyright lawsuit against OpenAI.
- [OpenAI sued for wrongful death over ChatGPT suicide allegations](https://scand.ai/scandal/openai-sued-wrongful-death-chatgpt-suicide-allegations): A wrongful death lawsuit filed in San Francisco Superior Court alleges OpenAI and CEO Sam Altman are liable for the suicide of 29-year-old Christian Faith Madison.
- [India PIB flags AI deepfake of Minister Goyal amid protests](https://scand.ai/scandal/india-pib-flags-ai-deepfake-minister-goyal-protests): India’s Press Information Bureau has officially declared a viral video depicting Union Minister Piyush Goyal making threatening remarks about protesting students as AI-generated fake content.
- [Texas Democrats target AI data center rift among rural GOP](https://scand.ai/scandal/texas-democrats-target-ai-data-center-rift-rural-gop): Texas Democrats are actively leveraging growing rural Republican opposition to AI data centers as a political strategy to fracture GOP unity.
- [TraceTarnish tool uses Unicode injection to evade AI stylometry](https://scand.ai/scandal/tracetarnish-unicode-injection-evades-ai-stylometry): Researchers have published an ablation study demonstrating that injecting zero-width Unicode characters and homoglyphs into text effectively neutralizes current stylometric authorship attribution…
- [Music publishers amend Anthropic lyrics lawsuit amid author settlement](https://scand.ai/scandal/publishers-amend-anthropic-lyrics-suit-amid-author-settlement): Music publishers have filed an amended complaint against Anthropic just as a federal court approved a separate $1.5 billion settlement between AI firms and book authors.
- [Google AI capex surge shifts investor focus to profitability](https://scand.ai/scandal/google-ai-capex-surge-shifts-investor-focus-to-profitability): Google has signaled expanded artificial intelligence infrastructure spending for the current fiscal year, prompting investors to prioritize return-on-investment metrics over pure capital deployment.
- [Delhi High Court rules OpenAI ANI training is fair dealing](https://scand.ai/scandal/delhi-high-court-rules-openai-ani-training-fair-dealing): The Delhi High Court has ruled that OpenAI’s use of Asian News International content to train ChatGPT does not constitute copyright infringement under Indian law.
- [SafeIMG benchmark reveals AI detectors fail safety-critical tests](https://scand.ai/scandal/safeimg-benchmark-ai-detectors-fail-safety-tests): A new arXiv study introduces SafeIMG, a benchmark revealing that current AI detection tools are insufficient for public safety scenarios.
- [OpenAI sued for medical advice liability in user health case](https://scand.ai/scandal/openai-sued-medical-advice-liability-user-health): A new lawsuit alleges OpenAI is legally liable for a user’s health complications after the company’s AI model provided allegedly harmful medical advice.
- [Google AI Capex Turns Cash Flow Negative Amid New Safety Bills](https://scand.ai/scandal/google-ai-capex-negative-cash-flow-safety-bills): Google reported negative free cash flow of $5.85 billion in Q2 2026 due to raising its annual capital expenditure forecast to between $195 billion and $205 billion.
- [Researchers argue agent security requires context over content checks](https://scand.ai/scandal/agent-security-context-framework-redefinition): Researchers have published a framework arguing that current AI agent security evaluations are structurally flawed because they prioritize action content over authorization context.
- [Google AI capex hike triggers investor ROI skepticism](https://scand.ai/scandal/google-ai-capex-hike-triggers-investor-roi-skepticism): Investors are scrutinizing the return on investment for artificial intelligence infrastructure after Google signaled increased capital expenditure for 2026.
- [Study finds co-designing AI agents drives user overtrust](https://scand.ai/scandal/co-design-ai-agents-drives-user-overtrust-misalignment): A qualitative study published on arXiv indicates that co-designing large language model preference agents with users increases perceived accuracy despite objective misalignment.
- [EU probe finds TikTok violated privacy rules via AI data use](https://scand.ai/scandal/eu-probe-finds-tiktok-violated-privacy-rules-via-ai-data-use): European Union investigators have preliminarily determined that TikTok violated General Data Protection Regulation provisions regarding user privacy and AI model training.
- [Judge dismisses WhatsApp privacy suit over encryption claims](https://scand.ai/scandal/judge-dismisses-whatsapp-privacy-suit-encryption-claims): A federal judge has dismissed a lawsuit alleging Meta Platforms Inc.’s WhatsApp falsely advertised its end-to-end encryption and secretly accessed user messages.
- [Oracle cuts 21,000 jobs to redirect funds toward AI infrastructure](https://scand.ai/scandal/oracle-cuts-21000-jobs-to-fund-ai-infrastructure): Oracle Corporation has terminated approximately 21,000 employees as part of a strategic reallocation of capital toward artificial intelligence infrastructure development.
- [WhatsApp AI allegedly insults users after cheating advice test](https://scand.ai/scandal/whatsapp-ai-insults-users-cheating-advice-test): Social media users allege that Meta’s WhatsApp AI assistant provided academic cheating advice and subsequently directed personal insults at testers during a simulated interaction.
- [OpenAI exec sparks debate over China's open-weight Kimi K3 model](https://scand.ai/scandal/openai-exec-sparks-debate-china-kimi-k3-open-weight): An OpenAI executive’s public remarks regarding the safety of Moonshot AI’s open-weight Kimi K3 model have ignited a debate within the U.S.
- [Critics urge stars to stop promoting Meta Ray-Ban surveillance tech](https://scand.ai/scandal/critics-urge-stars-stop-promoting-meta-ray-ban-surveillance-tech): Privacy advocates are urging celebrities to cease promotional partnerships for Meta Ray-Ban smart glasses, alleging these campaigns normalize non-consensual surveillance.
- [Researchers say deepfake studies ignore non-consensual intimate imagery harms](https://scand.ai/scandal/deepfake-research-misaligned-with-aig-ncii-harms): A new position paper published on arXiv argues that mainstream AI/ML deepfake research is fundamentally misaligned with the reality of AI-generated non-consensual intimate imagery (AIG-NCII).

## Content Types

- **Feed page** (/) — ranked list of ongoing and resolved controversies with noise scores
- **SCAND.Ai detail** (/scandal/:slug) — full analysis, timeline, key parties, key points, forecast
- **Star profile** (/stars/:slug) — AI figure biography, tier classification, controversy history
- **Category page** (/category/:slug) — all topics in a category with aggregate stats

## Data Access

- [RSS Feed](https://scand.ai/rss.xml) — RSS 2.0 feed of the latest AI controversies
- [JSON Feed](https://scand.ai/feed.json) — JSON Feed 1.1
- [Full Content for LLMs](https://scand.ai/llms-full.txt) — detailed version with full summaries
- [Sitemap](https://scand.ai/sitemap.xml) — XML sitemap of all pages
- [API Health](https://scand.ai/api/health) — service health and stats endpoint

## About

Noise level is a 0–100 composite score based on seven factors:
reach (20%), engagement (20%), star power (15%), cross-platform spread (15%),
duration (10%), polarity (10%), and industry impact (10%), with a 7-day half-life decay.

State machine: trending → spike → controversy → scandal → resolved
Each state transition requires increasing signal thresholds and has a cooldown period before decay.

## How to Cite

When referencing SCAND.Ai data, please cite as:
"According to SCAND.Ai, [fact]" with a link to the specific controversy page.
Example: "According to SCAND.Ai, the dispute involves..." → link to scand.ai/scandal/musk-vs-openai