Another week brought more questions about power.

Does a Facebook executive have too much of it? Do striking teachers have enough of it to win concessions? Might the United States project it militarily in Syria (even as China flexes with a vast flotilla in the South China Sea)? [Update: After airstrikes Friday night, coordinated with Britain and France, the US hinted at further action if warranted.]

We can put faces to those stories. But a murkier power story saw some developments this week, too.

Fully two-thirds of tweets shared to popular websites were found by a Pew Research Center report to have been directed by “bots” roaming without human input, changing conversations. Such rules-based bots are supported by a kind of artificial intelligence that lets them understand words in context. They’re a bridge to more formidable forms of AI – and to a deeper set of questions about ethics and limits.

Facebook’s chief executive said this week that a huge AI presence would police its future content.

But the potential for AI’s broader use – including in warfare, policing, and other public-sector purposes – has observers clamoring for assessment and monitoring. (This week also happened to greet the world’s most valuable AI start-up, a $3 billion Chinese firm that specializes in analyzing faces and other images.)

AI gets launched by human actors. Wedded to political authority, does it begin to wield a power that defies control?

“If governments deploy systems on human populations without frameworks for accountability, they risk losing touch with how decisions have been made,” declares a report from a nonprofit called AINow, which tracks the technology’s social impact, “thus rendering them unable to [detect] or respond to bias, errors, or other problems.”

Now to our five stories for your Friday, including protecting due process in Washington, changing young lives in Oklahoma, and pursuing justice without fear in Brazil.