The world’s leading publication for data science, AI, and ML professionals.

April Edition: Can AI Be Governed?

On the stakes and challenges of responsible innovation

Monthly Edition

It’s never been easy to distinguish real progress from hype when it comes to AI research, but we seem to have reached an inflection point in recent weeks. Tools that had previously required deep technical expertise have gone mainstream, and entire industries appear to be questioning their own chances of survival with the emergence of powerful—and accessible—AI applications.

Late last month, hundreds of notable researchers and tech leaders published a widely circulated petition calling to temporarily pause all development of AI systems that exceed GPT-4’s abilities. Many critical responses quickly followed, including an incisive dissent by "Stochastic Parrots" authors Timnit Gebru, Emily M. Bender, Angelina McMillan-Major, and Margaret Mitchell, who aimed to reframe the debate around what they see as the real ethical and social risks that AI development poses.

The broader conversation around the future of AI is complex, often confusing, and constantly shifting. To help our community stay informed of the wide range of perspectives on this topic, we’ve selected several standout articles that cover it from multiple angles. The authors of these posts eschew simplistic "AI: good or bad?" debates and add much-needed nuance to questions that have no easy answers.

Before we dive in, we wanted to thank you, as always, for all your support of the work we publish. For those of you who’d like to make a meaningful contribution, consider becoming Medium members; if you’re a student based in eligible countries, you can now sign up with a substantial discount.

TDS Editors


TDS Editors Highlights

  • Why I Signed the "Pause Giant AI Experiments" Petition (March 2023, 8 minutes) Despite being a signatory of the said petition, Rafe Brena, PhD has quite a few reservations about the way it approaches AI-related risks. His latest post unpacks the thinking process that led to his decision to add his name to the open letter after all, and describes the direction he hopes to see the industry move in.

  • Ethical Considerations In Machine Learning Projects (February 2023, 7 minutes) If you’re fairly new to the topic of AI ethics, Hennie de Harder‘s overview is a useful, beginner-friendly primer that covers some of the key concepts and challenges, including explainability and privacy.

  • What Is the EU’s Artificial Intelligence Act and What Will It Change? (August 2022, 16 minutes) One of the most ambitious attempts at regulating the creation and implementation of AI tools is the European Union’s Artificial Intelligence Act. Sara Tähtinen walks us through some of the key details in this proposed law, and offers a critical assessment of its strengths and shortcomings.

  • On AI Governance: How to Get Started (October 2022, 8 minutes) The responsible use of AI is crucial on the hyper-local level, too. Albert Suryadi argues that every organization needs to think about AI governance in order to build and maintain trust, and shares a roadmap for creating a robust internal structure to accomplish that goal.

  • Participatory Approaches to Algorithmic Responsibility (May 2022, 11 minutes) "In deliberative democracies," says Maya Murad, "citizens should have agency over how their data is used," including in the context of algorithmic decision-making systems (ADS) deployed by private entities. How to get there is tricky, but Maya suggests a participatory framework that enables people and organizations to become active partners in the process.

  • We Don’t Know How to Make AGI Safe (July 2022, 21 minutes) Discussions of artificial general intelligence—are AI researchers toying with disaster? Is it maybe just around the corner?—always run the risk of devolving into mediocre sci-fi (will the machines kill us?). Kyle O’Brien‘s reflection stands out precisely for acknowledging that risk and advocating for staying both level-headed and curious as AI enters uncharted territory.

Original Features

Explore our latest selection of resources and reading recommendations.


Popular Posts

In case you missed them, here are some of last month’s most-read posts on TDS.


We were thrilled to welcome a new cohort of TDS authors in March – they include João Pereira, Dorian Drost, Diogo Leitão, Leonardo Castorina, Jae Kim, Hang YU, Matt Chapman, and Robert Elmes, among others. If you have an interesting project or idea to share with us, we’d love to hear from you!

See you next month.


Related Articles