This week's edition explores Anthropic's new reasoning model, effective bias mitigation approaches, and insights on making better organizational decisions
🤖 Anthropic Launches Claude 3.7 Sonnet, First Hybrid Reasoning Model What it is: Anthropic, founded in 2021 by former OpenAI researchers, develops frontier AI models that combine powerful capabilities with advanced safety features. Their Claude series has gained popularity for its helpfulness, harmlessness, and honesty in conversational AI.
What's new: This week, Anthropic unveiled Claude 3.7 Sonnet, the first hybrid reasoning model that lets users choose between quick answers and deep, visible thinking. The system now allows you to see Claude's step-by-step reasoning process, making it easier to verify its work and learn from its approach. In benchmarks, Claude 3.7 Sonnet achieved remarkable results in coding tasks and complex problem-solving, particularly when given time to think. Anthropic also introduced "Claude Code," a command-line tool that can search codebases, write tests, and even commit directly to GitHub.
Why it matters: You can now use a single AI system for both quick everyday questions and complex problems requiring deep analysis, with added transparency into its process.
🧠 New Research Reveals Optimal Approaches to Mitigating Decision Bias What it is: A comprehensive study, co-authored by researchers from the London School of Economics, King's College London, and Bayes Business School, identifies two distinct approaches to improving decision-making in organizations: debiasing and choice architecture.
What's new: After analyzing 100 experimental studies, researchers identified when to use two distinct approaches to improve decision-making. "Debiasing" involves learning to recognize and counter your own biases through techniques like considering opposite viewpoints or statistical training. "Choice architecture" changes how information is presented to you—through defaults, visualization, or reframing—making better decisions more intuitive without requiring conscious effort.
Why it matters: Understanding which approach works best for different situations helps you make better decisions in both personal and professional contexts.
Citation: Fasolo, B., Heard, C., & Scopelliti, I. (2024). Mitigating Cognitive Bias to Improve Organizational Decisions: An Integrative Review, Framework, and Research Agenda. Journal of Management, 0(0). https://doi.org/10.1177/01492063241287188
More AI everywhere doesn’t mean better results. Small teams of AI experts beat mass licenses. Attackers now use AI to run cybercrime at machine speed. Leaders can’t keep up. Tools still forget us. The winners? Focused expertise and hybrid AI, not AI everywhere.
You've built something and you need to know if it works. So you do what's sensible—you ask an LLM to grade it. Factual accuracy, code quality, agent outputs. The machine judges the machine, and you get a number you can act on. Except that number
Imagine you have a bunch of teams, some with AI, some without, and some where everyone gets their own AI. Researchers ran a big experiment with over 400 people to see what actually happens when you mix and match humans and AI in different ways.
Here’s what they found:
