@micahbateman
Asst Prof in LIS. Digital humanities, cultural heritage/analytics, libraries/archives, poetry, Whitman. Finishing Lyric Publics: The Uses of Poetry in American Social-Media Campaigns. Opinions do not reflect employer. bit.ly/MicahBateman
New paper w/ @teddyroland.bsky.social on "How fiction powers generative AI systems." We designed a computational experiment to test the impact of the vast amount of fiction in LLM training data on how LLMs communicate, w/ implications for both AI design + literary theory. arxiv.org/abs/2603.01220
(This course came about from feedback that students are being placed between market imperatives to USE AI and instructor-level discipline-and-punish policies. But I also wanted to explore a corpus of literary works by Lillian-Yvonne Bertram, Ben Lerner, Vauhini Vara et al. who respond to gen AI.)
Major assessments consist of creative projects with critical process commentary, using writing as a method of inquiry into questions of AI, authorship, and textual production. Emphasis is placed on ethical use and professional judgment.
The course adopts a non-punitive, exploratory approach to AI engagement, emphasizing experimentation, iterative revision, and transparent reflection in order to support informed, critical decisions about when, how, and whether AI systems should be used.
Students engage scholarship from critical AI studies, authorship studies, and writing and textual history alongside contemporary literary and cultural works responding to AI or algorithmic writing tools.
A new course I'll be teaching Spring 2027
Authorship After Automation: Writing, Creativity, and Generative AI
This graduate course examines generative AI writing tools as information systems that produce, circulate, and transform text across creative, academic, and professional contexts. ...
My blood pressure wishes they hadn't.
The problem is I've written like AI my whole life. I love colons and em dashes and comparative "X rather than Y" framing and lists of three things.
Did the Greeks act upon an emotional belief in the gods >>>> Did the Greeks really believe in the gods
Lowkey tho I kinda thought we all knew this was happening as it was happening
You know what really sets the humanities back from taking a rigorous approach to AI? The state siccing AI on the humanities
MLA, ACLS, and AHA lawsuit reveals use of ChatGPT in illegal termination of grants by DOGE. Motion for summary judgment asserts violations of the First Amendment; violations of the Equal Protection Clause; and violation of the separation of powers. mla.org/NEH-Lawsuit
The MLA, @acls1919.bsky.social, and @historians.org have filed a motion for summary judgment in our lawsuit to restore the NEH. Discovery documents reveal that DOGE rather than the acting chair led grant terminations and targeted grants using ChatGPT. More at mla.org/NEH-Lawsuit
Huge NYT article on the back story of those two weeks in March/April 2025 when DOGE illegally terminated hundreds of NEH grants. Gift link!!
www.nytimes.com/2026/03/07/a...
Man I was really taking a break from subject-verb agreements in this thread.
This is part of what I'll contribute to the "AI in the Classroom" roundtable at the AmLitAssoc conference in Chicago. Hope to see some of you there!
And this cognitive process should transparently be a course learning objective.
And close reading as we ambiently understand it in the undergraduate classroom requires embodied and processual attention (the other half of the "Stop the Cop Shit" essay). AI restrictions should then be framed as protecting a cognitive process, not enforcing authorship or academic integrity.
The field can't agree on modes of reading, so courses form amorphously around the ambiently implied modes favored by the instructor. It's, I think, permissible to just state these outright to the students. Historical reading and close reading are fine if not exhaustive LOs.
Ironically the thing that can help you reverse-engineer learning objectives from your preexisting material is generative AI. But it's not necessary. The bigger obstacle to learning objectives in literary studies is methodological uncertainty.
This intentionality creates the more felicitous conditions for a 'community of inquiry' model of learning that has long served the discussion-oriented lit crit course. Whereas unscaffolded and often uninformed proscriptions against a ubiquitous technology creates a space of anxiety and distrust.
This creates a lack of transparency to the students who just experience proscriptions as disciplinary ("cop shit"), which erodes trust in the classroom and instructor and learning process. Whereas process-focused learning objectives that stress embodied pathways to discovery feels more intentional.
I think it's important for those banning AI in the classroom to separate disciplinary vs pedagogical AI proscriptions. It's hard to do this without learning objectives that align your pedagogy, and not many lit crit classrooms have learning objectives.
AI-generated artwork is officially not eligible for copyright protection after the U.S. Supreme Court declined to review the appeal. The court ruled that a work must have a human creator to qualify for copyright protection. www.reuters.com/legal/govern...
The provocative scenes were SO FLAT.
For instance, I am DEEPLY SCARED about what this means for the future of my own children. At the same time that I am SO EXCITED that it has made all my student assignments more accessible.
It still codes to me as "wake up!" But certainly there is exuberance in exploring this as a new and powerful tool. But that exuberance does not necessarily indicate a politics of unregulated AI evangelism.
I think the author's lane -- and PS this was generated by Claude -- is, "AI has changed social-science workflows, wake up!" I'm not seeing it as a net cheerleader for AI. Legislators', ethicists', activists' lane should be REGULATE REGULATE REGULATE.
As for individuals, you get to the same conundrum of, I shouldn't eat meat. I shouldn't drive a car. I shouldn't be on social media. I shouldn't use streaming. But, here we are.
R1 institutions are going full-force AI for a number of reasons but 1 is that they'd be fully left behind in grant money if they didn't. So that ain't happenin.
The kind of collective action that we can't seem to muster to do literally anything else.