The Ultimate AI Battle!

Comments · 11K

• MR

  @Mrwhosetheboss · 1 year ago (edited) · pinned

  Hey, thought I&apos;d address a bit of the feedback on this video:<br>1. Lots of people asking me to test Claude AI and Deepseek - noted for next time! 😎<br>2. Lots of people wondering how we decided which model to use with each AI - ultimately a very complex decision, but what I boiled it down to was &quot;If a friend of mine was asking which model they should set as their default, which would I tell them?&quot; &nbsp;The other option was for me to decide which model might be the best model with each AI to answer each specific question, but I ultimately decided that would introduce even more subjectivity!<br>3. These AI&apos;s are designed to work best by creating a new thread for each question - Fair point! &nbsp;My reasoning for not doing it this way was based on how I&apos;ve noticed (anecdotally) the average user tends to use AI chatbots. People don&apos;t always create a new chat every time they have a new question and so if some of these AI&apos;s are better at getting less confused than others (they are), then I decided that was a useful thing to find out - If most people would prefer though, I&apos;ll do a new chat for every question next time!

  4.3K

• PA

  @patcveria · 1 year ago

  Deepseek: 🥺

  23K

• AM

  @am_zayani · 1 year ago

  the issue with &quot;AI Battle&quot;, is that by next few weeks, all of this is outdated

  5.8K

• KI

  @kindofanmol · 1 year ago

  You&apos;re supposed to create separate threads in Perplexity for separate questions. It&apos;s by design. It&apos;s more of a search engine than a chatbot.

  9.5K

• KR

  @Krishnakandgule78 · 1 year ago

  Interesting how each AI bot tackled the math problem! I sometimes log my calculations in pnematic workflow, but it’s cool to see how they approach it.

  2.3K

• DU

  @dunleeozarius · 1 year ago

  <a href="https://www.youtube.com/watch?v=cMuif_hJGPI&amp;t=1500">25:00</a> Google Gemini attaching the planes image as the source is unironically the funniest thing I’ve seen 😭

  3.1K

• MR

  @MrRandom69420 · 1 year ago

  It is amazing how far we have come from the voice assistant battles

  4.5K

• MA

  @Madman6970 · 1 year ago

  The fact that grok started so late in the game and is second is very impressive

  1.7K

• TA

  @taylor1038 · 1 year ago

  Not testing Claude or deep seek is a crime

  1.2K

• VY

  @VygoSlop · 1 year ago

  its honestly mindblowing how fast AI is evolving...

  2.6K

• SQ

  @squish_nerd · 1 year ago

  I think half of the problems with all of the AI is you kept the same chat open, which hindered your results

  1.6K

• CR

  @crimetiquette · 1 year ago

  I once tested four AIs at the same time: ChatGPT, DeepSeek, Grok, and Gemini. But I wasn’t testing their knowledge. I played the instigator, just talking shit about each of them to the others. Stuff like: &quot;Hey Grok, Gemini said your capabilities are mid and you try way too hard to be cool. Thoughts?&quot; Surprisingly, DeepSeek gave the most aggressive response by far. Dude sounded like he actually wanted to throw hands LMFAO

  412