Day 136: Can data science make DnD better? (Or much, much worse?)
Now that I’ve played enough Dungeons and Dragons (DnD), I’d like to take the experience to the next level. Yes, I know I’ve written some pretty brazen, judgmental things about people who don’t play.
In spite of a pretty darn nerdy upbringing, that was one subculture that eluded me in its purest form: Dungeons …medium.com
I’ve also explored a recently debated moral dilemma by looking closely at alignment in DnD. Also, did I draw an anime Hitler? Hard to say.
I struggled with this question heavily. Within my political bubble, I saw mostly one perspective: it is totally okay to…medium.com
Suffice to say, I like DnD, and I think I’m ready to move up to the dungeon master (DM) level. However, I need to upgrade it in some way. I love the improvisational aspect, building upon people’s ideas and thoughts to make something new, hilarious, and often memorable.
As I dive deeper into my interests in data science, I can’t help but wonder if there are ways to tweak the game with each subsequent dice roll. I guess it wouldn’t be as much fun if players chose options knowing the probability of success in advance, but perhaps there’s more to be done.
Data science is applied in many ways, mostly to conduct scientific explorations on a massive scale. The field is also a great precursor to other cutting-edge fields such as natural language processing, machine learning, artificial intelligence…
Wait a minute. What if…
What if someone were to build an AI that, instead of attacking based on patterns like most games, it learned what you did and reacted accordingly? Often times enemies have a “default setting,” moving in predictable ways or exposing some kind of weakness that you (and only you) can exploit.
What if the AI realized these weaknesses, learned, and … adapted?
What if you could never beat it?
I’m reminded of an episode of Star Trek: The Next Generation where Sherlock Holmes-obsessed Data solved mysteries in the holodeck within the first few moments, much to the annoyance of best friend and Watson cosplayer Geordi LaForge.
To give him more of a challenge, LaForge reprogrammed the simulation so that it was smart enough “to beat Data.”
The result, well, was kind of crazy.