The field of artificial intelligence is moving towards a more nuanced understanding of AI capabilities and their impact on market valuations. Researchers are developing new frameworks to evaluate AI systems, focusing on validity and transparency, and proposing methods to quantify the gap between AI potential and realized performance. Embodied AI is also gaining momentum, with a growing recognition of the need for more accurate quality indicators for robots operating in real-world environments. Noteworthy papers include:

• Measurement to Meaning: A Validity-Centered Framework for AI Evaluation, which proposes a structured approach to evaluating AI systems.
• On the Evaluation of Engineering Artificial General Intelligence, which advances the state of the art in benchmarking and evaluation of AI agents in engineering design contexts.
• Toward Embodied AGI: A Review of Embodied AI and the Road Ahead, which introduces a systematic taxonomy of Embodied AGI and proposes a conceptual framework for a robotic brain.