Rebecca Crootof

Describe something you learned from lecture this week that you didn’t know just from doing the readings. Connect something you learned in 209 this week to something you are learning in another of your classes this semester. If Monday and Wednesday assigned readings were from two different texts, make connections between the two. Do they share themes? Forms? Tone? Historical context? Do you find them equally interesting? Look at the very first paragraph of one of our texts and discuss how the opening lays the groundwork for the rest of the work. Quote specific lines, phrases, or images. Using quotes from a text, persuade someone (friend or foe, your grandma or your senator) to change their mind about something important. Compose a 5 song playlist to accompany an assigned reading from this week, and write a few sentences for each song, explaining your choices. Describe an idea you had in response to the readings/lecture/discussions for our class this week- any idea, about literature, or the world, or yourself. How might you pursue this idea, in your studies or elsewhere? Choose a passage (no more than 10 lines) from the readings this week and rewrite it, changing at least one of the literary aspects such as: person (change from first to third-person or vice versa), tense (change from past to present etc), focalizing character (i.e. write it from a different character's perspective), style (adjectives, diction, description, tone). Then write 2-3 sentences about the effect of your changes. Compose a yelp review to strangers, or a letter to a specific person, or a booktok style video, recommending a novel/poem/play from this week’s reading. Write a letter to someone who has questioned your choice of majoring or minoring in English, explaining why you value what you’re learning. Include quotes / ideas from this week’s readings.

syllabus time, teaming up for the herculean efforts of reinventing writing assignments. here, some prompts for required weekly low stakes 250-500 word reflections/ process pieces that have proven relatively conducive to real writing in lit class. please share any similar suggestions in thread.

i keep coming back to this.

Anyone working on inserting secret "sleeper" viruses in their novels that can be activated only inside LLMs? Perhaps along the lines of Snow Crash.

Cumulative Civilian Harm in Gaza: A Gendered View We need to reconceive of the consequences of violence in Gaza for civilians as composite, aggregate, collective, and layered harms.

Current measures of #CivilianHarm do not adequately account for the aggregate impacts of protracted armed conflict on civilian populations, exemplified by the war in #Gaza.

Fionnuala Ni Aolain offers the concept of cumulative civilian harm as an analytic tool:

www.justsecurity.org/115407/cumul...

Tech companies only have five ideas: robot slave (actually just human slaves), hallucinatory counter-reality, untaxable money, The Everything App, and Clippy

everyone in the natsec and FP community increasingly just losing their minds as the topical questions in US FP and natsec these days increasingly devolve to the level of "should we lick this toaster while it is on? some experts weigh in"

Army Eliminates Office for Minimizing Civilian Deaths on Battlefields The Civilian Protection Center of Excellence, which once housed roughly 30 staff, has been folded into the Army's dense web of unrelated bureaucratic policy shops.

A huge step backwards. To save nearly no money, in military terms, we incur such a deep cost.

Military AI as Sociotechnical Systems - Lieber Institute West Point Conceptualizing weapons as sociotechnical systems enables us to focus on risks arising with any incorporation of AI into the use of a weapon.

Nightstand Pile Files: Military AI as Sociotechnical Systems by Mitt Regan

Mitt Regan talking about how miliary AI decision assistants should be regulated as sociotechnical systems. What's not to love???

A great intro for those new to any of these issues or teaching an "AI & Law" class.

19/18:
@haaretzcom.bsky.social archived here without paywall.
“Enough is Enough. Israel is committing war crimes”
by ex-Likud conservative former Prime Minister Ehud Olmert:

archive.is/2025.05.28-1...

Chain-Gang All-Stars Two top women gladiators fight for their freedom within…

Nightstand Pile Files: Chain-Gang All-Stars by Nana Kwame Adjei-Brenyah

Wow. Hunger Games meets systemic racism/the carceral state/abolitionism . . . with FOOTNOTES. With a dash of unintended side effects of well-intentioned research.

And distinctive, understandable, deeply human characters.

Soviet officer who averted cold war nuclear disaster dies aged 77 ‘Gut instinct’ told Lt Col Stanislav Petrov that apparent launch of US missiles was actually early warning system malfunction

Stanislav Petrov, the man who saved the world, died on this day in 2017. If you don’t know the story, just read this www.theguardian.com/world/2017/s...

Happy to! Feel free to reach out to my richmond.edu account.

My quibble is that this proposal is focused solely on atrocity victims, and I would like to expand it to encompass all victims of wartime harms, both lawful and unlawful.

But every movement needs to start somewhere, and atrocity victims unquestionably deserve more than they're now receiving.

Nightstand Pile Files: Transnational Atrocity Victim Networking: A New Paradigm for International Accountability Mechanisms @profgsgordon.bsky.social

A creative proposal for connecting and empowering victims of wartime harms, to facilitate participation and compensation.

Implementing #WarTorts

Red teaming is great for identifying one source of risk - namely, malicious acts and the accident risks that they foreshadow.

But absent intentional consideration of likely harms that arise when everything works WELL and tech proliferates, red teaming won't anticipate all likely risks and impacts.

Huge !!

[Then, once "self-regulating" leads to all kinds of negative societal fall-out . . .]

(2) If we must have regulations, let's have industry-informed regulations. After all, we're the experts. And you don't want to stifle innovation!

Not sure that these are other "stages" but @bjard.bsky.social and I often refer to industry's "regulatory two-step":

(1) We have incentives to self-regulate! And you don't want to stifle innovation!

Truth.

U.S. Army Plans to Eliminate Office for Reducing Civilian Harm in War

Also worth noting subsequent news:

Analyzes the respective benefits of 'but for' and proximate causation in this context, with sensitivity to what each would require of the commander.

Unsurprisingly, this ultimately reinforces my long-held stance that the state - not individuals - should be responsible for wartime civilian harms.

The U.S. DoD’s New Civilian Harm Mitigation and Response Policy and “Harm Resulting From” Military Operations - Lieber Institute West Point The new U.S. DoDI is an important development in the conduct of U.S. military operations and the reach of LOAC more generally.

Nightstand Pile Files: The U.S. DoD's New CHMR Policy and "Harm Resulting From" Military Operations @danmaurerlawprof.bsky.social

The DoD definition of "civilian harm" has the qualifier "resulting from military operations" - what causation test should be used to evaluate that?

#WarTorts

the fundamental disconnect between the manosphere and reality is that they are incapable of understanding that this is - according to every woman I’ve ever met - the sexiest thing a man has ever done

The thought experiment was right: we have invented a paperclip machine.

Truly impressive investigative work.

Indeed difficult, but vital to watch.

In addition to @janinadill.bsky.social's analysis, a few LOAC notes:

--no claim of warning prior to firing on ambulances
--commander's failure to punish can also be war crime
--IDF personnel may face universal jx abroad

it's useful to think that all science had to be funded into existence and prior to its modern professionalization, almost all science was carried out by the idle rich (bros)

that is, the science that exists is the science that people wanted to exist, and a lot of the time it's to argue something

AI Red-Teaming Is Not a One-Stop Solution to AI Harms: Recommendations for Using Red-Teaming for AI Accountability Increasingly, red-teaming is being put forward as a solution to concerns about artificial intelligence — a way to pressure test AI systems and identify potential harms. What does that mean in practice...

Nightstand Pile Files: AI Red-Teaming is Not a One-Stop Solution to AI Harms @friedler.net Ranji Singh @borhane.bsky.social Jacob Metcalf Brian J. Chen

Admittedly, I've been too often defaulting to "Wouldn't red teaming address that?" This is a concise clarifier of when (and how) it can - or won't.

Meta’s ‘Digital Companions’ Will Talk Sex With Users—Even Children Chatbots on Instagram, Facebook and WhatsApp are empowered to engage in ‘romantic role-play’ that can turn explicit. Some people inside the company are concerned.

Fearing that Meta’s chatbots were too boring, Mark Zuckerberg ordered they be given more human-like traits. That rebuke led to a loosening of boundaries -- and the launch of digital companions that Meta knew could produce sex role-play content for children.
www.wsj.com/tech/ai/meta...

Nation Can’t Believe It On Harvard’s Side

Nation Can’t Believe It On Harvard’s Side

I’m sorry but if your students use AI to write papers and you use AI to grade them zero school is happening. You are running together on a hamster wheel