EthicAI, a University of Cambridge spinout operating in ethical and responsible assurance of AI, has launched a cloud-based AI assurance SaaS platform.
The launch of BeehAIve aims to solve the problems traditionally created by AI models such as fairness and bias, transparency and impact on employees.
BeehAIve scrutinises models and datasets against 15 ethical dimensions, assessing evidence across thousands of measurement points.
The launch also allows for technical stress-testing of model performance to improve reliability, robustness and security while recommending ways to reduce energy usage from AI.
The platform enables assessment of models and systems against regional and national legislation including the EU AI Act and GDPR.
The system then recommends how gaps can be filled, preparing for cross-border compliance while building stakeholder trust.
Tanya Goodin, chief executive and founder of EthicAI, said: “Research from Edelman shows that consumer confidence in AI has fallen by 8% globally over the past five years due to concerns about how AI is designed and built.
“BeehAIve looks under the hood of organisations’ AI models and systems – both in development and post deployment – to identify blind spots and weaknesses and provides actionable insights on how to mitigate and manage risks.”
The post Cambridge spinout launches new tool to boost confidence in AI appeared first on UKTN.
