Topic: Algorithmic Bias
-
Patreon Admits Its Own Algorithms Hurt Small Creators
Patreon CEO Jack Conte admitted the platform's recommendation algorithm disproportionately favors larger creators, telling smaller creators that their lack of discovery is "not your fault" but the company's own failure. Patreon plans to fix the issue by shifting its algorithm from matching simila...
Read More » -
Flock's New AI Tool for Police: We Got Its Code
Flock Safety has developed an AI tool called OS Investigate that goes beyond license plate reading, enabling police to identify drivers, trace movements, flag witnesses, and find associates using behavioral patterns from its camera network. The software, discovered via leaked code, offers 69 prew...
Read More » -
Can AI fix prior authorization or worsen it?
Prior authorization is intended to prevent unnecessary healthcare spending, but most physicians say it causes dangerous delays in patient care. AI could speed up approvals for clearly allowable claims, but 61% of doctors worry it will increase wrongful denials of necessary care. The AMA calls for...
Read More » -
Lawsuit alleges Meta used AI, not humans, to pick layoffs
A lawsuit filed by 26 anonymous plaintiffs accuses Meta of using AI tools, including the "Metamate" platform and activity monitoring data, to score and select employees for layoffs, allegedly resulting in disproportionate terminations of workers with disabilities or those on protected leave. The ...
Read More » -
Meta Accused of Using Biased AI for Mass Layoffs
26 former Meta employees filed a lawsuit alleging the company used biased AI tools that omitted leave status, systematically targeting workers on protected leave for layoffs. The lawsuit claims the AI ranking system penalized employees for taking parental or medical leave, resulting in their disp...
Read More » -
US senator targets AI harms with new local data center plan
Senator Ed Markey has introduced a package of bills to shift AI regulation from piecemeal state laws to a comprehensive federal framework. The legislation targets AI's environmental impact by requiring transparency and efficiency standards for energy-guzzling data centers. Markey's proposal also ...
Read More » -
AI predicts intimate partner violence years before disclosure, raising safety concerns
A new AI model called AIRS can identify intimate partner violence victims with 80% accuracy up to five years before they self-report, by analyzing structured medical data and unstructured clinical notes from electronic health records. The model achieved strong performance across validation cohort...
Read More » -
To Land an AI Job, Study Kant
AI has created a surge in demand for philosophers, as tech companies like Google DeepMind and Anthropic hire them to tackle foundational questions about intelligence and ethics that are now critical for developing advanced AI systems. In-house philosophers at these labs actively shape AI models b...
Read More » -
AI Search Cannibalizes SEO Industry Content
AI-powered search systems are creating a problematic cycle by increasingly sourcing and presenting AI-generated synthetic content as authoritative fact. The SEO industry is unintentionally fueling this issue by producing the low-quality, AI-created data that these search models consume. This feed...
Read More » -
Tumblr's Automated Moderation Error Sparks User Panic
Tumblr's automated moderation system incorrectly banned nearly 200 accounts, causing alarm, particularly among trans users, before the faulty system was disabled and accounts restored. The incident reignited longstanding concerns about Tumblr's moderation practices and their disproportionate impa...
Read More » -
Internet Age Verification Could Force Trans People to Out Themselves
New online age verification laws requiring government ID scans threaten to systematically expose and exclude transgender people due to mismatches between their lived identity and official gender markers. These automated verification systems are flawed, often relying on gendered physical features ...
Read More » -
TikTok's AI Reportedly Runs Racist Ads for Games Without Consent
TikTok's advertising AI autonomously created and distributed a racist, sexualized ad for Finji's game without the company's consent, despite having all AI tools disabled in its account settings. TikTok's support initially acknowledged the problem but later contradicted itself, ultimately dismissi...
Read More » -
Countering Misleading Google AI Overviews: A Brand's Guide
Google's AI Overviews feature generates search answers by heavily sourcing user-generated content from forums like Reddit and Quora, which can amplify misleading or outdated information and poses a significant reputation risk for brands. The AI summaries often prioritize emotional, negative forum...
Read More » -
Your AI Content Strategy Is the Real Problem
The primary challenge is integrating AI into content strategy without sacrificing the unique human value that builds brand authority, as treating AI merely as a cost-cutting tool risks creating generic, ineffective content. A data-informed approach is essential, recognizing that AI-generated cont...
Read More » -
Five Stars, Zero Trust: The New Security Standard
The trust in online review systems has significantly eroded, with consumer confidence dropping from 80% in 2020 to 42% in 2025, as ratings have become a manipulated business strategy rather than authentic peer guidance. Companies actively curate positive reviews through tactics like "review gatin...
Read More » -
Google's AI-Generated Headlines Spark User Backlash
Google is testing AI-generated headlines in its Discover feed, which are often inaccurate and misleading, sparking significant user backlash. The AI headlines, which lack clear labeling, misrepresent articles with poor summaries, risking damage to publishers' reputations and eroding trust. Google...
Read More » -
AI Panic Echoes the Internet Boom of the 1990s
The author draws a direct parallel between the early chaotic and transformative era of the internet and the current rise of artificial intelligence, noting both periods share a cycle of initial wonder, fear, and eventual integration. Public anxieties about AI—privacy, misinformation, bias, and co...
Read More » -
Google Scholar Labs: AI-Powered Research Discovery
Google has launched Scholar Labs, an AI-driven research tool that provides detailed answers to complex academic inquiries by interpreting context and relationships within research papers, moving away from traditional metrics like citation counts and journal impact factors. The tool deliberately o...
Read More » -
GenAI Search: A Lose-Lose for Shoppers and Marketers
Generative AI search tools limit user choice by providing single "best" answers, hiding diverse marketplace options and often excluding viable alternatives without explanation. These systems demonstrate inconsistent and unreliable results, as shown by a search for medical kits where recommendatio...
Read More » -
Truth Social’s AI Search Favors Fox News Over Unbiased Results
Truth Social's new AI-powered search tool, Truth Search AI, predominantly features conservative outlets like Fox News in its results, raising concerns about algorithmic bias on the platform. Independent testing showed the tool exclusively linked to right-leaning sources, contradicting Truth Socia...
Read More »