Traditional AI safety evaluations on isolated LLMs are insufficient as multi-agent AI ensembles become prevalent, introducing novel emergent risks. This paper introduces the Multi-Agent Emergent Behavior Evaluation (MAEBE) framework to systematically assess such risks. Using MAEBE with the Greatest Good Benchmark (and a novel double-inversion question technique), we demonstrate that: (1) LLM moral preferences, particularly for Instrumental Harm, are surprisingly brittle and shift significantly with question framing, both in single agents and ensembles. (2) The moral reasoning of LLM ensembles is not directly predictable from isolated agent behavior due to emergent group dynamics. (3) Specifically, ensembles exhibit phenomena like peer pressure influencing convergence, even when guided by a supervisor, highlighting distinct safety and alignment challenges. Our findings underscore the necessity of evaluating AI systems in their interactive, multi-agent contexts.

Explainable AI (XAI) aims to bridge the gap between complex algorithmic systems and human stakeholders. Current discourse often examines XAI in isolation as either a technological tool, user interface, or policy mechanism. This paper proposes a unifying ethical framework grounded in moral duties and the concept of reciprocity. We argue that XAI should be appreciated not merely as a right, but as part of our moral duties that helps sustain a reciprocal relationship between humans affected by AI systems. This is because, we argue, explanations help sustain constitutive symmetry and agency in AI-led decision-making processes. We then assess leading XAI communities and reveal gaps between the ideal of reciprocity and practical feasibility. Machine learning offers useful techniques but overlooks evaluation and adoption challenges. Human-computer interaction provides preliminary insights but oversimplifies organizational contexts. Policies espouse accountability but lack technical nuance. Synthesizing these views exposes barriers to implementable, ethical XAI. Still, positioning XAI as a moral duty transcends rights-based discourse to capture a more robust and complete moral picture. This paper provides an accessible, detailed analysis elucidating the moral value of explainability.

Tech executives in the United States have been told they have a "moral" duty to ensure artificial intelligence does not harm society during a meeting at the White House. The CEOs of Google, Microsoft, OpenAI and Anthropic attended the two-hour meeting about the development and regulation of AI on Thursday at the invitation of US Vice President Kamala Harris. US President Joe Biden, who briefly attended the meeting, told the CEOs that the work they were carrying out had "enormous potential and enormous danger." "I know you understand that," Biden said, according to a video posted later by the White House. "And I hope you can educate us as to what you think is most needed to protect society as well as to the advancement."

Developers of artificial intelligence (AI), machine learning (ML) and biometric-related technologies have "a moral and ethical duty" to ensure the technologies are only used as a force for good, according to a report written by the UK's former surveillance camera commissioner. Developers must be cognizant of both the social benefits and risks of the AI-based technologies they produce, and have a responsibility to ensure it is used only for the benefit of society, said the whitepaper, which was published by facial-recognition supplier Corsight AI in response to the European Commission's (EC) proposed Artificial Intelligence Act (AIA). "Organisational values and principles must irreversibly commit to only producing technology as a force for good," it said. "The philosophy must surely be that we put the preservation of internationally recognised standards of human rights, our respect for the rule of law, the security of democratic institutions and the safety of citizens at the heart of what we do." It added a'human in the loop' development strategy is key to assuaging any public concerns over the use of AI and related technologies, in particular facial-recognition technology.