tkgally

16 ms·

1. 181.▲
   by tkgally 1y ago

   > ... a translators’ and interpreters’ work is mostly about ensuring context, navigating ambiguity, and handling cultural sensitivity. This is what Google Translate cannot currently do. Google Translate can't, but LLMs given enough
2. 182.▲
   by tkgally 1y ago

   The results are not surprising to me personally. When I have used AI to help with my own writing and translation tasks, I do not feel as mentally engaged with the writing or translation process as I would be if I were doing it all on my own
3. 183.▲
   by tkgally 1y ago

   The points made in the abstract of that paper look like pretty serious challenges to the Apple paper. What might be the counterobjections?
4. 184.▲
   by tkgally 1y ago

   When you’re done talking, you tap anywhere on the screen to indicate that you’re finished talking, and Claude replies a second or two later. If you are silent for several seconds, it goes ahead and replies without waiting for you to tap. Wh
5. 185.▲
   by tkgally 1y ago

   Both off and on topic: I got access to Claude’s voice mode today, and for my second test I asked it for its thoughts about both the linked blog post and the HN discussion here. The full transcript is at [1]. Some parts that stood out (sligh
6. 186.▲
   by tkgally 1y ago

   From the author's note at the end: "An earlier version of this article was submitted to the U.S. Army War College’s War Room blog back in the fall of 2024.... The piece was accepted and scheduled for publication in February. I fou
7. 187.▲
   Failure Mechanisms in Democratic Regimes – An Army's Role(angrystaffofficer.com)

   126 points by tkgally 1y ago|97 comments
8. 188.▲
   by tkgally 1y ago

   Thanks for that recommendation. Though on a different topic, I found the following post particularly interesting: "Failure Mechanisms in Democratic Regimes – an Army’s Role" https://angrystaffofficer.com/2025/
9. 189.▲
   by tkgally 1y ago

   And from the bullet points and the lack of a personal perspective. I'm glad the poster at least admitted the AI contribution, though. I use AI a lot myself for brainstorming and perspective and even advice. But I include in my prompt d
10. 190.▲
    by tkgally 1y ago

    Long-time Japan resident here. The IC cards do work quickly and smoothly, but the retail payment system overall is a mess because different stores accept different combinations of dozens of electronic payment brands. When shopping, I prefer
11. 191.▲
    'We Are the Most Rejected Generation'(nytimes.com)

    4 points by tkgally 1y ago|0 comments
12. 192.▲
    by tkgally 1y ago

    I understand. I don’t get exactly the same degree of roasting each time. The roasted beans themselves are not as consistently roasted as ideal, either; after I grind them, I notice some lighter and darker grounds mixed in. But that doesn’t
13. 193.▲
    by tkgally 1y ago

    I started roasting my own coffee a few months ago. I use a light pan with a screen cover I found online and roast the beans by shaking the pan over a high flame for seven or eight minutes. I judge the degree of the roast from the sound, the
14. 194.▲
    America's Coming Brain Drain(foreignaffairs.com)

    7 points by tkgally 1y ago|7 comments
15. 195.▲
    by tkgally 1y ago

    I teach a university class in which I ask the students to submit writing each week, and I have also seen obviously LLM-produced writing. Yes, it’s boring and doesn’t show the students’ thinking, and the students are not getting any wiser by
16. 196.▲
    by tkgally 1y ago

    For the past couple of months, I’ve been running occasional side-by-side tests of the deep research products from OpenAI, Google, Perplexity, DeepSeek, and others. Ever since Google upgraded its deep research model to Gemini 2.5 Pro Experim
17. 197.▲
    by tkgally 1y ago

    The link is here: https://notebooklm.google.com/notebook/c36ea335-6686-474d-bf... The fumu fumu is at 01:50. The podcast is about the impact of AI on higher education in Japan. I prompted NotebookLM briefly in Japane
18. 198.▲
    by tkgally 1y ago

    Yes. In fact, I laughed when I heard the NotebookLM voice say that. It was comically out of place in the context.
19. 199.▲
    by tkgally 1y ago

    I tried it with Japanese, and it sounded about as good as in English. Only at one point did it sound unnatural. Japanese two-person conversation uses a lot of backchannelling ( aizuchi ), that is, semilinguistic sounds made by the listener
20. 200.▲
    by tkgally 1y ago

    I tried it with Japanese, and it sounded about as good as in English. Only at one point did it sound unnatural. Japanese two-person conversation uses a lot of backchannelling, that is, semilinguistic sounds made by the listener ( aizuchi ).
21. 201.▲
    by tkgally 1y ago

    The video discusses that issue briefly: https://youtu.be/Lo0FDmSbTp4?si=hu1E6phsUiazZ2YH&t=1879
22. 202.▲
    "OpenAI Is Not God" – The DeepSeek Documentary [video](youtube.com)

    6 points by tkgally 1y ago|2 comments
23. 203.▲
    by tkgally 1y ago

    I teach at a university in Japan, and, for the past two and a half years, I have been struggling with the implications of AI for university education. I found this essay interesting and helpful. One remark: > I fed the entire nine-hundre
24. 204.▲
    Will the Humanities Survive Artificial Intelligence?(newyorker.com)

    45 points by tkgally 1y ago|68 comments
25. 205.▲
    by tkgally 1y ago

    Calvin Trillin is one of my favorite writers. The nonfiction—both serious and humorous—he has written for the New Yorker since the 1960s is particularly good.
26. 206.▲
    by tkgally 2y ago

    Still later: I later read that o3 is supposedly particularly good with this geoguessing, so I tried the same prompt and photos with o3. This time it got one out of four correct: “The view of the canal with cherry blossoms and the green rail
27. 207.▲
    by tkgally 2y ago

    I asked the just-released ChatGPT o4-mini-high to locate four photographs of varying difficulty. It didn’t get any of them right, though the guesses weren’t bad. The reasoning was also interesting to watch, as it cropped sections of the pho
28. 208.▲
    by tkgally 2y ago

    Another useful word in this context is “sycophancy,” meaning excessive flattery or insincere agreement. Amanda Askell of Anthropic has used it to describe a trait they try to suppress in Claude: https://youtube.com/watch?v=u
29. 209.▲
    by tkgally 2y ago

    Four-decade resident of Japan here. While small towns might have their appeal, for me the real Japan is the Yamanote Line between Shinjuku and Ikebukuro on a Friday evening in summer, the cars packed with sweaty drunk and sober people of al
30. 210.▲
    by tkgally 2y ago

    > My prompts are falling behind.... If you haven’t already, you might want to try metaprompting, that is, having a model write the prompt for you. These days, I usually dictate my metaprompts through a STT app, which saves me a lot of ti