Episode 5 of AI Ethics with Fexingo tackles a next-generation bias problem: training data generated by AI itself. Lucas explains how synthetic data — used to save costs and protect privacy — can amplify existing prejudices when models learn from their own outputs. The conversation centers on one concrete case: a medical imaging model that inherited skin-tone bias from a synthetic dataset derived from a real dataset that already underrepresented darker skin. Lucas and Luna walk through the feedback loop, why it's hard to detect, and what regulators like the EU AI Act are starting to require. A specific, sobering look at how well-intentioned shortcuts in AI development can quietly make bias worse.

#SyntheticData #AIbias #Ethics #MachineLearning #MedicalAI #FeedbackLoops #DataQuality #EUAIAct #Representation #ModelCollapse #AIAccountability #AlgorithmicFairness #HealthTech #TechRegulation #ResponsibleAI #FexingoBusiness #BusinessPodcast #Technology

Keep every episode free: buymeacoffee.com/fexingo

Episode 4 of AI Ethics with Fexingo dives into the European Union's Artificial Intelligence Act, the first comprehensive legal framework for AI. Lucas and Luna explore how the Act classifies AI systems by risk, what it means for companies like OpenAI and Google, and whether it could set a global precedent similar to GDPR. They discuss the specific prohibitions on social scoring and real-time biometric surveillance, the obligations for high-risk systems like medical devices and hiring tools, and the potential challenges of enforcement. The episode also touches on the political compromises that watered down provisions on facial recognition and the ongoing debate over foundation models. With comparisons to GDPR's global ripple effect, the hosts ask whether the AI Act will become a blueprint or a bureaucratic burden. A concrete look at regulation that will affect anyone building or using AI in Europe and beyond.

#EUAIAct #ArtificialIntelligence #AIPolicy #GDPR #TechRegulation #BrusselsEffect #FoundationModels #FacialRecognition #BiometricSurveillance #HighRiskAI #OpenAI #Google #DeepMind #EuropeanCommission #DigitalRights #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo

Lucas and Luna examine a real-world case where a widely used healthcare algorithm, designed to manage chronic disease care, systematically underestimated the health needs of Black patients. Data from a 2019 Science study shows the algorithm was 50% less likely to flag Black patients for extra care compared to white patients with the same chronic conditions. They break down how a flawed proxy—healthcare spending rather than actual illness—introduced racial bias, why the hospital network initially missed the problem, and the practical fix the researchers proposed. The episode also explores why this kind of bias is harder to spot in healthcare than in hiring or lending algorithms. No jargon, just a clear look at how a well-intentioned system went wrong and what it means for the future of AI-driven medicine.

#AlgorithmicBias #HealthcareAI #RacialDisparities #ChronicDisease #PredictiveAlgorithms #MachineLearning #HealthEquity #ScienceStudy #BlackPatients #BiasInMedicine #AIEthics #ClinicalDecisionSupport #DataProxy #AlgorithmAudit #Technology #FexingoBusiness #BusinessPodcast #AI

Keep every episode free: buymeacoffee.com/fexingo

Every year, millions of job applications are filtered by algorithms that claim to be objective. But a 2023 study of Amazon's now-defunct AI recruiting tool revealed something disturbing: the system systematically downgraded resumes containing the word 'women's' — as in 'women's chess club captain' or 'women's soccer league president.' Lucas and Luna unpack how that bias crept in, why it wasn't caught until after deployment, and what it means for the hundreds of companies now using AI to screen candidates. They explore the 2018 Reuters investigation that first flagged the problem, the internal Amazon emails that showed engineers knew about the bias but shipped anyway, and the broader lesson: AI ethics isn't just a philosophy problem — it's an engineering problem with real people on the losing end. No hand-waving. Just the data and the decisions behind one of the most consequential AI failures of the past decade.

#AIEthics #RecruitmentBias #AmazonAI #AlgorithmicBias #WomenInTech #JobSearch #HiringAlgorithms #BiasInAI #ResponsibleAI #AIRegulation #TechEthics #ReutersInvestigation #GenderBias #HRTech #AIAccountability #FexingoBusiness #BusinessPodcast #Technology

Keep every episode free: buymeacoffee.com/fexingo

In the premiere episode of AI Ethics with Fexingo, Lucas and Luna launch the show by drilling into a single case: the COMPAS recidivism algorithm used in U.S. courtrooms. They walk through the 2016 ProPublica investigation that found the tool was nearly twice as likely to falsely flag Black defendants as future criminals than white defendants — while also being less accurate overall than a random person on the internet guessing. Lucas explains how COMPAS worked, why the error rates mattered in real sentencing decisions, and why the algorithm's 'black box' design made it impossible for defendants to challenge. Luna pushes back on whether the problem is the algorithm or the system it was built for, and both hosts outline the show's promise: not hand-wringing about hypothetical robot uprisings, but the concrete, often boring decisions in code that determine whether a tool helps or harms. No hot takes. Just the numbers, the tradeoffs, and the people affected.

#AIEthics #COMPAS #ProPublica #CriminalJustice #AlgorithmicBias #Recidivism #PredictivePolicing #JuliaAngwin #Northpointe #EqualProtection #Sentencing #FalsePositive #Fairness #Accountability #Transparency #Technology #FexingoBusiness #BusinessPodcast

Keep every episode free: buymeacoffee.com/fexingo