I was invited to deliver a keynote speech at the ‘AI for Good Summit’ this year, and I arrived at the venue with an open mind and hope for change. With a title “AI for social good: the new face of technosolutionism” and an abstract that clearly outlined the need to question what “good” is and the importance of confronting power, it wouldn’t be difficult to guess what my keynote planned to address. I had hoped my invitation to the summit was the beginning of engaging in critical self-reflection for the community. 

But this is what happened. Two hours before I was to deliver my keynote, the organisers approached me without prior warning and informed me that they had flagged my talk and it needed substantial altering or that I would have to withdraw myself as speaker. I had submitted the abstract for my talk to the summit over a month before, clearly indicating the kind of topics I planned to cover. I also submitted the slides for my talk a week prior to the event. 

Thinking that it would be better to deliver some of my message than none, I went through the charade or reviewing my slide deck with them, being told to remove any reference to “Gaza” or “Palestine” or “Israel” and editing the word “genocide” to “war crimes” until only a single slide that called for “No AI for War Crimes” remained. That is where I drew the line. I was then told that  even displaying that slide was not acceptable and I had to withdraw, a decision they reversed about 10 minutes later, shortly before I took to the stage.

Looking at this year’s keynote and centre stage speakers, an overwhelming number of them came from industry, including Meta, Microsoft, and Amazon. Out of the 82 centre stage speakers, 37 came from industry, compared to five from academia and only three from civil society organisations. This shows that what “good” means in the "AI for Good" summit is overwhelmingly shaped, defined, and actively curated by the tech industry, which holds a vested interest in societal uptake of AI regardless of any risk or harm.

“AI for Good”, but good for whom and for what? Good PR for big tech corporations? Good for laundering accountability? Good for the atrocities the AI industry is aiding and abetting? Good for boosting the very technologies that are widening inequity, destroying the environment, and concentrating power and resources in the hands of few? Good for AI acceleration completely void of any critical thinking about its societal implications? Good for jumping on the next AI trend regardless of its merit, usefulness, or functionality? Good for displaying and promoting commercial products and parading robots?

Any ‘AI for Good’ initiative that serves as a stage that platforms big tech, while censoring anyone that dares to point out the industry’s complacency in enabling and powering genocide and other atrocity crimes is also complicit. For a United Nations Summit whose brand is founded upon doing good, to pressure a Black woman academic to curb her critique of powerful corporations should make it clear that the summit is only good for the industry. And that it is business, not people, that counts.

This is a condensed, edited version of a blog Abeba Birhane published earlier this month. The conference organisers, the International Telecommunication Union, a UN agency, said “all speakers are welcome to share their personal viewpoints about the role of technology in society” but it did not deny demanding cuts to Birhane’s talk. Birhane told Coda that “no one from the ITU or the Summit has reached out” and “no apologies have been issued so far.”

A version of this story was published in the Coda Currents newsletter. Sign up here.

The post AI, the UN and the performance of virtue  appeared first on Coda Story.