Auditing Algorithmic Risk
Cathy O’Neil, Jake Appel, and Sam Tyner-Monroe
Key Insight: A set of frameworks can help organizations identify potential harms posed by algorithms, AI tools, or large language models (LLMs).
Top Takeaways: As business uses of artificial intelligence, LLMs, and other algorithmic applications expand, so, too, does the possibility of unintended negative consequences for users and other stakeholders. The authors introduce two auditing frameworks — Ethical Matrix and Explainable Fairness — that can help organizations identify these potential risks and address critical questions about who could be harmed by algorithmic systems and how. They also discuss applying red teaming and benchmarking to difficult-to-audit LLMs, before applying all four of the approaches to a real-life example to demonstrate how an algorithmic audit could have prevented a very public chatbot failure.
Avoid ML Failures by Asking the Right Questions
Dusan Popovic, Shreyas Lakhtakia, Will Landecker, and Melissa Valentine
Key Insight: Checking assumptions and mapping out work processes can help ensure that machine learning solutions fit the job to be done.
Top Takeaways: A significant reason why machine learning projects fail to deliver business value is data scientists’ failure to adequately understand the business context. Development teams can avoid mistakes when they put aside any reticence to ask basic questions and engage with colleagues on the business side. The authors advise gaining input from all involved stakeholders and suggest some specific types of queries that might help machine learning developers get to the heart of the problem at hand.
How Generative AI Can Support Advanced Analytics Practice
Pedro Amorim and João Alves
Key Insight: The natural language capabilities of large language models (LLMs) can augment the predictive powers of advanced analytics when well-designed prompts are applied.
Top Takeaways: Advanced analytics, such as predictive and prescriptive models to support business decisions, remain the primary drivers of data science value in the enterprise. How might the flashy, fluent, but not entirely reliable generative AI LLMs contribute to traditional analytics practice? The authors describe some experimental prompts that show potential for labeling data and explaining model predictions, and share guidance on monitoring and verifying that output.
Managing Data Privacy Risk in Advanced Analytics
Gregory Vial, Julien Crowe, and Patrick Mesana
Key Insight: Gaining value from data assets that include customers’ personal information requires the use of privacy techniques that balance data utility with data protection.
Top Takeaways: Many companies have large stores of customer data that can be tapped for valuable insights via analytics. At the same time, the cybersecurity tactics for protecting personal information within that data can render it less useful for analysis. Data science practices will increasingly require that teams collaborate with IT on each use case to identify which techniques will maximize data privacy while still exposing useful information in the data set for analysis. Organizations can achieve better balance between data utility and data security by including data privacy in data literacy programs, treating data privacy as a business issue, and formalizing their approach to addressing the issue.
Acing Value-Based Sales
Marco Bertini, Oded Koenigsberg, and Todd Snelgrove
Key Insight: Selling value requires more than quantifying benefits. It demands a deep commitment to collaborating with customers.
Top Takeaways: Value-based selling efforts often fizzle after an initial push because companies fail to see beyond the numbers when calculating the economic impact of product or service benefits. While quantifying evidence of benefits is at the heart of value-based selling, it’s not enough. Companies can follow a five-step process that establishes a basis for joint value creation with customers and enables them to establish a deep mutual understanding of how both the buyer and seller can benefit from and share in that value.
Find a Circular Strategy to Fit Your Business Model
Samsurin Welch and Khaled Soufani
Key Insight: Analyzing your products’ full life cycles can enable your organization to extract more value from materials and resources, and improve both sustainability and profitability.
Top Takeaways: Circular business models consider the full lifespan of a product, taking into account how materials are sourced; how the product is manufactured, distributed, and used; and what happens to the product at its end of life. When companies operate in this way, they stand to boost their own efficiency and competitiveness while better meeting their sustainability goals. The authors describe four types of circular business models — extending product lifetimes, tapping idle or wasted capacity, reclaiming material resources, and using resources more frugally — and discuss the capabilities required to execute each approach.
How to Come Back Stronger From Organizational Trauma
Payal Sharma
Key Insight: Leaders play a key role in promoting the growth and development that can emerge after trauma.
Top Takeaways: Traumatic events in the workplace, whether through acts of violence or natural phenomena, leave individuals and organizations reeling and destabilized. However, psychology research has identified a phenomenon called post-traumatic growth (PTG), whereby such experiences can enable survivors to develop new capabilities and a more nuanced understanding of the world. In this article, the author looks at how leaders can apply the concepts behind PTG to help their organizations move forward after a traumatic incident and help employees rebuild their sense of safety, control, protection, and purpose.
Engineer Your Own Luck
Mark J. Greeven, Howard Yu, and Jialu Shan
Key Insight: No one can predict the future, but modularizing core capabilities can position companies to be ready for the unexpected.
Top Takeaways: Businesses that can move quickly to capture new opportunities don’t have greater foresight than the rest. What they do have, increasingly, is optionality: the ability to discover and act on their options. One way to maximize an organization’s options for growth is by treating internal capabilities as plug-and-play, mix-and-match modular digital services. This approach can enable businesses to scale quickly, and help their partners and clients grow new market segments.
Serve More Customers With Inclusive Product Design
Vanessa M. Patrick and Jeffrey D. Shulman
Key Insight: Centering on the most marginalized users in the product development process can result in features that benefit all customers.
Top Takeaways: Products designed for the average user can unintentionally exclude those with special needs or from marginalized populations. Considering diversity along multiple dimensions — including age, disability, religion, and nationality — can help product teams widen their view of their target market. Even small increases in attentiveness to underserved users’ needs can yield significant results by making products usable by a larger population. Pilot projects can help a team analyze how this kind of attentive engagement will affect product development, including time and financial costs and product performance.
The CEO’s Cyber Resilience Playbook
Manuel Hepfer, Rashmy Chatterjee, and Michael Smets
Key Insight: CEOs are critical to orchestrating a shift from simply playing cybersecurity defense to developing an action plan for handling a cyberattack when — not if — it occurs.
Top Takeaways: Many organizations over-rely on cybersecurity defenses at the expense of building cyber resilience. CEOs whose companies have been through a malware attack or hacking incident stress the importance of accepting that cyberattacks are inevitable and making business continuity and recovery plans long before they’re needed. Drawing on their conversations with such CEOs, the authors share best practices for building an organization’s cyber resilience and managing stakeholders during these incidents.
“The MIT Sloan Management Review is a research-based magazine and digital platform for business executives published at the MIT Sloan School of Management.”
Post details > link to site
