France: CNAF's discriminatory risk-scoring algorithm must be stopped
Amnesty International and coalition partners filed a complaint with the Council of State against CNAF's risk-scoring algorithm used to detect benefit overpayments. The algorithm assigns risk scores based on criteria that discriminate against vulnerable groups, including those with disabilities, single parents, and low-income households. The complaint alleges the system violates human rights to equality and privacy. The EU AI Act's social scoring ban may apply, but its definition remains unclear.
- Company involved
- CNAF
2 source articles · read the reporting →
Texas AG Settles with Pieces Technologies Over Deceptive Healthcare AI Claims
The Texas Attorney General investigated Pieces Technologies, a Dallas-based healthcare AI company, for making false and misleading statements about the accuracy of its generative AI product used in hospitals. The company claimed an error rate of less than 1 per 100,000, but the investigation found these metrics were likely inaccurate. As part of the settlement, Pieces agreed to accurately disclose its product's accuracy and ensure hospital staff understand the appropriate reliance on its AI. The case marks the first-of-its-kind healthcare generative AI investigation by the AG.
- Company involved
- Pieces Technologies
4 source articles · read the reporting →
KPMG drops AI report after false case studies exposed
KPMG removed a global report on AI after multiple case studies were found to be inaccurate and apparently generated from AI hallucinations. The report falsely claimed that UBS, NHS Greater Manchester, Swiss Federal Railways and Transport for London used AI agents in various ways. The companies denied the claims, and KPMG took the report down while reviewing its production.
- Company involved
- KPMG
3 source articles · read the reporting →
Fraudsters use AI voice deepfake to trick UK energy firm CEO into transferring $243,000
In March 2019, criminals used commercially available AI voice-generation software to impersonate the CEO of a German parent company. They tricked the CEO of a UK-based energy firm into urgently wiring $243,000 to a Hungarian supplier. The fraud was discovered when the fraudsters attempted a second transfer, which the CEO refused. The company was insured and the loss was covered.
- Company involved
- Unnamed UK-based energy firm
- AI system involved
- Commercially available voice-generating AI software
10 source articles · read the reporting →
Walmart employees claim Everseen anti-theft AI is flawed
A group of Walmart employees, calling themselves the Concerned Home Office Associates, alleged that the Everseen AI system used in thousands of stores to prevent shoplifting frequently produces false positives, flagging innocent customer behaviour as theft. They created a video showing the system failing to stop actual theft and mistakenly alerting for a phone. Employees said the false positives frustrate customers and lead to unnecessary interventions, raising public health concerns during the COVID-19 pandemic. Walmart denied a widespread issue but said it had made improvements.
- Company involved
- Walmart
- AI system involved
- Everseen
9 source articles · read the reporting →
UIUC researchers use OpenAI API to automate phone scams for under a dollar
Researchers at the University of Illinois Urbana-Champaign used OpenAI's Realtime API to create AI agents that can autonomously execute phone scams. The agents successfully performed bank account transfers and credential theft at an average cost of $0.75 per scam. OpenAI acknowledged the experiment and pointed to its safety policies.
- Company involved
- University of Illinois Urbana-Champaign
- AI system involved
- GPT-4o Realtime API
6 source articles · read the reporting →
Epic sepsis prediction algorithm misses two-thirds of cases, study finds
A study of nearly 30,000 patients at University of Michigan hospitals found that Epic Systems' sepsis prediction algorithm missed two-thirds of sepsis cases and frequently issued false alarms. The algorithm, used by over 100 US health systems, is designed to alert staff to potential sepsis, but the study suggests it provides little value and may contribute to alert fatigue. Epic disputed the study's conclusions, claiming its system has helped save lives.
- Company involved
- Epic Systems
- AI system involved
- Epic sepsis prediction model
6 source articles · read the reporting →
Shipt's new V2 algorithm reduces pay for 41% of workers, study finds
A study by MIT Media Lab and Coworker.org found that Shipt's new V2 payment algorithm reduced pay for 41% of workers by an average of 11% per shop. The algorithm, rolled out in September 2020, replaced a transparent payment system with a black-box model. Workers pooled data to analyze the impact, revealing uneven distribution of earnings.
- Company involved
- Shipt
- AI system involved
- V2
10 source articles · read the reporting →
Google AI Overviews provide inaccurate finance information in 43% of searches
A study by The College Investor found that Google's AI Overviews provided misleading or inaccurate information in 43% of 100 personal finance-related searches. The AI-generated answers included outdated tax rules and incorrect student loan repayment plan details. One user believed she could convert her 529 plan to a Roth IRA in California, which is not allowed. Google has not responded to requests for comment.
- Company involved
- Google
- AI system involved
- AI Overviews
3 source articles · read the reporting →
Wisconsin’s dropout prediction algorithm labels students high risk with racial bias
Wisconsin’s Dropout Early Warning System (DEWS) uses machine learning to predict middle school students' likelihood of graduating on time, relying on factors including race and income. The Markup found the algorithm frequently mislabels students, with a higher false alarm rate for Black and Hispanic students. The Wisconsin Department of Public Instruction has not informed schools of the racial disparities and continues to use the system.
- Company involved
- Wisconsin Department of Public Instruction
- AI system involved
- Dropout Early Warning System (DEWS)
10 source articles · read the reporting →
DeepScore markets facial and voice analysis app for trustworthiness scoring despite experts' doubts
DeepScore, a Tokyo-based company, is marketing an app that uses facial and voice recognition to score people's trustworthiness for lenders and insurers in Japan, Indonesia, Vietnam and the Philippines. The company says the app can detect deception with 70 per cent accuracy, but researchers and privacy advocates say there is no reliable scientific basis for such judgments and warn of discrimination and privacy harms. The chief executive said the system is only one part of lenders' and insurers' decision-making and that people can choose not to use it. Critics respond that an unequal balance of power makes consent difficult.
- Company involved
- DeepScore
- AI system involved
- DeepScore
6 source articles · read the reporting →
Hospitals use Epic AI to predict Covid-19 decline without validation
Dozens of hospitals across the US are using Epic's deterioration index AI system to predict which Covid-19 patients will become critically ill, despite the tool not being validated for the new disease. The rapid deployment during the pandemic bypassed normal testing and validation processes.
- AI system involved
- Deterioration index
9 source articles · read the reporting →
Google and HCA Healthcare partner to analyze 32 million patient records
Google Cloud has announced a partnership with HCA Healthcare to analyze around 32 million patient records. The anonymized data will be used to develop algorithms that could advise doctors on treatment options. Privacy advocates have raised concerns about the data sharing and the potential for AI to re-identify patients. HCA insists that patient-identifiable information will be stripped and that access will be tightly controlled.
- Company involved
- HCA Healthcare
- AI system involved
- Google Cloud
9 source articles · read the reporting →
University of Chicago report finds US healthcare algorithms rife with bias
A University of Chicago report alleges that algorithms used by hospitals, insurers and other businesses across the US are biased along racial and economic lines. The algorithms help triage patients, predict diabetes and flag those needing extra care, but the report says flawed products were introduced with little oversight. It says inequitable treatment has continued for more than a decade.
7 source articles · read the reporting →
Denmark's Udbetaling Danmark welfare algorithm illegally surveilled millions
Udbetaling Danmark (UDK), Denmark's centralized welfare payment agency, used automated algorithms to check eligibility and detect fraud. The system collected personal data on millions of beneficiaries and their relatives, including non-beneficiaries, in what a think-tank called 'systematic surveillance'. A complaint to the Danish Data Protection Authority led to a ruling that the data collection violated GDPR, and UDK eventually deleted the illegally obtained data. The system also caused erroneous payments, including underpaying 300,000 pensioners by 94 million euros annually.
- Company involved
- Udbetaling Danmark
10 source articles · read the reporting →
California regulator accuses Maxpread Technologies of using AI-generated fake CEO to scam investors
Maxpread Technologies allegedly used an AI-generated avatar to pose as its CEO in a YouTube video to deceive investors. The California DFPI issued cease and desist letters to Maxpread and four other companies for offering unqualified securities and making false promises. The regulator claims the companies promised high daily returns using AI trading, but these claims were false. Maxpread did not respond to requests for comment.
- Company involved
- Maxpread Technologies
8 source articles · read the reporting →
Study finds credit score algorithms less accurate for minorities
A study of 50 million US consumers found that credit scoring algorithms used by mortgage lenders are less accurate for minority and low-income applicants due to sparse credit data, leading to higher rejection rates. The inaccuracy is due to noise in the data, not bias, so fairer algorithms cannot fix it. The study suggests that adjusting for bias had no effect, and that addressing the inaccuracy could reduce disparities by 50%.
- Company involved
- Mortgage lenders (unnamed)
- AI system involved
- Credit scoring algorithms
6 source articles · read the reporting →
UnitedHealthcare sued for using AI to deny Medicare Advantage patients care
A class action complaint alleges that UnitedHealthcare and its subsidiary naviHealth deployed an artificial intelligence model to override treating physicians' determinations of medically necessary care for elderly Medicare Advantage patients. The complaint claims the AI model has a known 90% error rate and wrongfully denied patients care. The lawsuit, filed in the U.S. District Court for the District of Minnesota, seeks damages on behalf of the estates of two deceased patients and all others similarly situated.
- Company involved
- UnitedHealthcare, Inc.
10 source articles · read the reporting →
Google Health's diabetic retinopathy AI faced real-world issues in Thailand clinics
Google Health deployed a deep-learning system to screen for diabetic retinopathy in 11 clinics across Thailand. The AI was highly accurate in lab tests but rejected over a fifth of eye scans in real-world conditions due to poor lighting and slow internet. Patients whose scans were rejected had to visit specialists at other clinics, causing inconvenience and frustration for nurses. Google Health is now working with local staff to improve the system's workflow.
- Company involved
- Google Health
10 source articles · read the reporting →
UnitedHealth used ALERT algorithm to limit mental health coverage, regulators found
ProPublica reports that UnitedHealth Group's Optum subsidiary used the ALERT algorithm to flag mental health patients and therapists as outliers, leading to therapy coverage denials. Regulators in California, Massachusetts and New York alleged this breached the federal Mental Health Parity and Addiction Equity Act, and UnitedHealth agreed to restrict the system in those jurisdictions. The company denies wrongdoing and says its programmes are compliant. Affected patients, including Medicaid enrollees, were left to pay out-of-pocket or go without care.
- Company involved
- UnitedHealth Group
- AI system involved
- ALERT
5 source articles · read the reporting →
Four commercial large language models perpetuate race-based medical misconceptions
A study published in npj Digital Medicine tested four commercial large language models (Bard, ChatGPT, GPT-4, and Claude) for their tendency to propagate discredited race-based medical beliefs. When asked about kidney function, lung capacity, and skin thickness, the models sometimes endorsed debunked racial differences, particularly affecting Black patients. The study concludes that these biases pose a potential hazard and urges caution before using such models in clinical decision-making.
- Company involved
- Not named in article (refers to commercial LLMs generically as Google's Bard, OpenAI's ChatGPT and GPT-4, and Anthropic's Claude)
- AI system involved
- Bard, ChatGPT, GPT-4, Claude
6 source articles · read the reporting →
Amazon merchants complain about AI review summaries focusing on negatives
Amazon introduced an AI-powered feature to generate summaries of customer product reviews. Merchants reported that the AI inaccurately highlights negative feedback, even when only a small percentage of reviews are critical. The summaries may exaggerate negative themes, potentially harming sales. Amazon acknowledged the issue and said it is working to refine the technology based on seller feedback.
- Company involved
- Amazon
- AI system involved
- AI-powered review highlights
5 source articles · read the reporting →
Humana sued for using AI to deny seniors rehabilitation care
Health insurer Humana is accused in a class-action lawsuit of using an AI algorithm to systematically deny rehabilitation care to Medicare Advantage patients, despite recommendations from their doctors. The lawsuit, filed on December 12, 2023, alleges that the AI tool restricted medically necessary care. This is the second major health insurer to face legal action over its use of AI to deny care.
- Company involved
- Humana
9 source articles · read the reporting →
Study finds ChatGPT provides inaccurate drug information responses
A study presented at the ASHP Midyear Clinical Meeting found that ChatGPT's responses to nearly three-quarters of drug-related questions were incomplete or inaccurate. The AI system also generated fake citations to support some responses. Researchers warned that healthcare professionals and patients should verify ChatGPT's medication information using trusted sources to avoid potential harm.
- Company involved
- OpenAI
- AI system involved
- ChatGPT
8 source articles · read the reporting →
