AI toxicity arises from complex socio-technical interactions among data, model architectures, and feedback loops, not solely from faulty training data.
If Architectures Could Talk, They'd Quote Your Boss
Software architecture mirrors organizational communication and incentives; failures stem from unclear ownership, misaligned incentives, and social friction rather than purely technical issues.
