Mark Riedl
@markriedl.bsky.social
AI for storytelling, games, explainability, safety, ethics. Professor at Georgia Tech. Associate Director of ML Center at GT. Time travel expert. Geek. Dad. he/him
I'm flying next week. I've just discovered the Flight Aware misery map www.flightaware.com/miserymap/
November 8, 2025 at 2:08 PM
I'm flying next week. I've just discovered the Flight Aware misery map www.flightaware.com/miserymap/
It’s turkey season (in the US). Gemini wants you to know that a wattle is very, very fleshy
November 6, 2025 at 9:12 PM
It’s turkey season (in the US). Gemini wants you to know that a wattle is very, very fleshy
Do you work on computational journalism, or computing in civic life, communication, public policy, political science, statistics, or law?
You may be interested in our Bharat Endowed Faculty Professorships at the Georgia Tech College of Computing www.cc.gatech.edu/bharat-endow...
You may be interested in our Bharat Endowed Faculty Professorships at the Georgia Tech College of Computing www.cc.gatech.edu/bharat-endow...
November 6, 2025 at 2:18 PM
Do you work on computational journalism, or computing in civic life, communication, public policy, political science, statistics, or law?
You may be interested in our Bharat Endowed Faculty Professorships at the Georgia Tech College of Computing www.cc.gatech.edu/bharat-endow...
You may be interested in our Bharat Endowed Faculty Professorships at the Georgia Tech College of Computing www.cc.gatech.edu/bharat-endow...
Here at the AI Ethics and Society Conference, I’m best known as “ @upolehsan.bsky.social ‘s Advisor”
October 21, 2025 at 1:25 PM
Here at the AI Ethics and Society Conference, I’m best known as “ @upolehsan.bsky.social ‘s Advisor”
I'm on my way to the AAAI Conference on AI, Ethics, and Society to present on "AI Agents and the Law".
What does the law of human agency tell us about liability for AI agents? We are missing some important things when it comes to alignment post-training.
arxiv.org/abs/2508.08544
See you there!
What does the law of human agency tell us about liability for AI agents? We are missing some important things when it comes to alignment post-training.
arxiv.org/abs/2508.08544
See you there!
October 18, 2025 at 9:59 PM
I'm on my way to the AAAI Conference on AI, Ethics, and Society to present on "AI Agents and the Law".
What does the law of human agency tell us about liability for AI agents? We are missing some important things when it comes to alignment post-training.
arxiv.org/abs/2508.08544
See you there!
What does the law of human agency tell us about liability for AI agents? We are missing some important things when it comes to alignment post-training.
arxiv.org/abs/2508.08544
See you there!
The 2nd Summit on Responsible Computing, AI, and Society kicks off Oct 28 and 29 rcais.github.io
Our agenda will feature: education, creativity, health, the humanities, sustainability, and aligned and accountable agents
We will be streaming most talks. Registration is open for remote attendees.
Our agenda will feature: education, creativity, health, the humanities, sustainability, and aligned and accountable agents
We will be streaming most talks. Registration is open for remote attendees.
October 16, 2025 at 2:18 PM
The 2nd Summit on Responsible Computing, AI, and Society kicks off Oct 28 and 29 rcais.github.io
Our agenda will feature: education, creativity, health, the humanities, sustainability, and aligned and accountable agents
We will be streaming most talks. Registration is open for remote attendees.
Our agenda will feature: education, creativity, health, the humanities, sustainability, and aligned and accountable agents
We will be streaming most talks. Registration is open for remote attendees.
It looks like my phone is trying to learn my commute. But my regular commute is by bike along bike paths that mostly don’t follow roads. The AI is trying to map my data to roads cars can drive and the result is hilarious.
October 14, 2025 at 9:23 PM
It looks like my phone is trying to learn my commute. But my regular commute is by bike along bike paths that mostly don’t follow roads. The AI is trying to map my data to roads cars can drive and the result is hilarious.
Hi all,
Georgia Tech is holding its 2nd Summit on Responsible Computing, AI, and Society October 28-29th in Atlanta.
rcais.github.io
Our speakers and schedule is set. It's going to be great!
If you are planning to be in Atlanta and want to join us, registration is now open on the website.
Georgia Tech is holding its 2nd Summit on Responsible Computing, AI, and Society October 28-29th in Atlanta.
rcais.github.io
Our speakers and schedule is set. It's going to be great!
If you are planning to be in Atlanta and want to join us, registration is now open on the website.
October 8, 2025 at 1:35 AM
Hi all,
Georgia Tech is holding its 2nd Summit on Responsible Computing, AI, and Society October 28-29th in Atlanta.
rcais.github.io
Our speakers and schedule is set. It's going to be great!
If you are planning to be in Atlanta and want to join us, registration is now open on the website.
Georgia Tech is holding its 2nd Summit on Responsible Computing, AI, and Society October 28-29th in Atlanta.
rcais.github.io
Our speakers and schedule is set. It's going to be great!
If you are planning to be in Atlanta and want to join us, registration is now open on the website.
If you're at #COLM2025 on Friday Oct 10, I'll be giving an invited talk in the Workshop on the Application of LLM Explainability to Reasoning and Planning.
The title is: "Reasoning, Explanation, and Human-Centered AI".
Below I have tentatively diagrammed the flow of my talk.
The title is: "Reasoning, Explanation, and Human-Centered AI".
Below I have tentatively diagrammed the flow of my talk.
October 6, 2025 at 6:53 PM
If you're at #COLM2025 on Friday Oct 10, I'll be giving an invited talk in the Workshop on the Application of LLM Explainability to Reasoning and Planning.
The title is: "Reasoning, Explanation, and Human-Centered AI".
Below I have tentatively diagrammed the flow of my talk.
The title is: "Reasoning, Explanation, and Human-Centered AI".
Below I have tentatively diagrammed the flow of my talk.
My local bookstore is amazing littleshopofstories.com
Live D&D session featuring local authors @runwithskizzers.bsky.social @jacksonpearce.bsky.social @markdoesstuff.bsky.social. They were great. Like having Critical Role in one’s backyard
Live D&D session featuring local authors @runwithskizzers.bsky.social @jacksonpearce.bsky.social @markdoesstuff.bsky.social. They were great. Like having Critical Role in one’s backyard
October 5, 2025 at 10:53 PM
My local bookstore is amazing littleshopofstories.com
Live D&D session featuring local authors @runwithskizzers.bsky.social @jacksonpearce.bsky.social @markdoesstuff.bsky.social. They were great. Like having Critical Role in one’s backyard
Live D&D session featuring local authors @runwithskizzers.bsky.social @jacksonpearce.bsky.social @markdoesstuff.bsky.social. They were great. Like having Critical Role in one’s backyard
Well that was quite the rabbit hole. tfwiki.net/wiki/Reprodu...
October 3, 2025 at 6:46 PM
Well that was quite the rabbit hole. tfwiki.net/wiki/Reprodu...
We observe a rise in AI for games that can be classified as wargames, regardless of domain.
Games with open-ended player actions and open-ended adjudication are the hardest for AI. LLMs are most applicable here. But it requires careful thought or human decision-making can be negatively impacted.
Games with open-ended player actions and open-ended adjudication are the hardest for AI. LLMs are most applicable here. But it requires careful thought or human decision-making can be negatively impacted.
September 30, 2025 at 4:57 PM
We observe a rise in AI for games that can be classified as wargames, regardless of domain.
Games with open-ended player actions and open-ended adjudication are the hardest for AI. LLMs are most applicable here. But it requires careful thought or human decision-making can be negatively impacted.
Games with open-ended player actions and open-ended adjudication are the hardest for AI. LLMs are most applicable here. But it requires careful thought or human decision-making can be negatively impacted.
Wargames have a long history, from Chess to Dungeons & Dragons. The characteristics of the game determine how we approach them with AI.
We provide an ontology based on how open-ended player actions can be, as well as how open-ended adjudication (game/dungeon mastering) can be.
We provide an ontology based on how open-ended player actions can be, as well as how open-ended adjudication (game/dungeon mastering) can be.
September 30, 2025 at 4:57 PM
Wargames have a long history, from Chess to Dungeons & Dragons. The characteristics of the game determine how we approach them with AI.
We provide an ontology based on how open-ended player actions can be, as well as how open-ended adjudication (game/dungeon mastering) can be.
We provide an ontology based on how open-ended player actions can be, as well as how open-ended adjudication (game/dungeon mastering) can be.
"Would you like to play a game?"
"Wargames" are simulations involving adversarial conflict that are used to assist in decision-making.
More than war, they are used for business, geopolitics, emergency response, entertainment, & more.
We wrote a paper on LLMs for wargames
arxiv.org/abs/2509.17192
"Wargames" are simulations involving adversarial conflict that are used to assist in decision-making.
More than war, they are used for business, geopolitics, emergency response, entertainment, & more.
We wrote a paper on LLMs for wargames
arxiv.org/abs/2509.17192
September 30, 2025 at 4:57 PM
"Would you like to play a game?"
"Wargames" are simulations involving adversarial conflict that are used to assist in decision-making.
More than war, they are used for business, geopolitics, emergency response, entertainment, & more.
We wrote a paper on LLMs for wargames
arxiv.org/abs/2509.17192
"Wargames" are simulations involving adversarial conflict that are used to assist in decision-making.
More than war, they are used for business, geopolitics, emergency response, entertainment, & more.
We wrote a paper on LLMs for wargames
arxiv.org/abs/2509.17192
My TA was making AI-generated solutions to test for cheating and got this result from ChatGPT.
Is ChatGPT now trained to refuse to do school work? Anyone else seeing things like this?
Is ChatGPT now trained to refuse to do school work? Anyone else seeing things like this?
September 22, 2025 at 11:08 PM
My TA was making AI-generated solutions to test for cheating and got this result from ChatGPT.
Is ChatGPT now trained to refuse to do school work? Anyone else seeing things like this?
Is ChatGPT now trained to refuse to do school work? Anyone else seeing things like this?
Some people watched Blade Runner and thought: what a great idea
September 19, 2025 at 1:55 PM
Some people watched Blade Runner and thought: what a great idea
Yes, Typst, these are all very good and useful suggestions, thank you.
September 13, 2025 at 1:19 PM
Yes, Typst, these are all very good and useful suggestions, thank you.
One day I will see a random person at the airport chatting away with the OG chatbot, Eliza. I am still searching.
September 12, 2025 at 10:54 AM
One day I will see a random person at the airport chatting away with the OG chatbot, Eliza. I am still searching.
September 10, 2025 at 3:30 PM
OpenAI put out a technical report on “Why Language Models Hallucinate”
It is is collaboration with Georgia Tech colleague Santosh Vempala
cdn.openai.com/pdf/d04913be...
It is is collaboration with Georgia Tech colleague Santosh Vempala
cdn.openai.com/pdf/d04913be...
September 6, 2025 at 3:55 PM
OpenAI put out a technical report on “Why Language Models Hallucinate”
It is is collaboration with Georgia Tech colleague Santosh Vempala
cdn.openai.com/pdf/d04913be...
It is is collaboration with Georgia Tech colleague Santosh Vempala
cdn.openai.com/pdf/d04913be...
Yes, but has /your/ neighborhood pizza joint rebranded around Star Trek?
September 5, 2025 at 11:14 PM
Yes, but has /your/ neighborhood pizza joint rebranded around Star Trek?
In which I opine about AI-generated bigfoot vlogs news.gatech.edu/news/2025/08...
September 4, 2025 at 3:05 PM
In which I opine about AI-generated bigfoot vlogs news.gatech.edu/news/2025/08...
OpenAI is discovering what every social media company has also discovered: content moderation is hard and AI content moderation is also hard.
September 4, 2025 at 1:48 AM
OpenAI is discovering what every social media company has also discovered: content moderation is hard and AI content moderation is also hard.