iiinova — where your digital self keeps evolving

Thomas Winters @thomaswinters.be · 22/01/2026

In het "TorfsBot or Not?" experiment konden mensen maar 70% van de tijd het juiste onderscheid maken tussen Rik Torfs tweets vs een autocomplete model getraind op zijn teksten.

230 3

Thomas Winters @thomaswinters.be · 11/10/2025

Steeds meer mensen gebruiken AI voor therapie. Maar praat ChatGPT je niet te veel naar de mond? Mijn comments hierover in De Wereld van Sofie (rond 1h03m30) www.vrt.be/vrtmax/luist...

vrt.be

VRT MAX

030 3

Thomas Winters @thomaswinters.be · 30/09/2025

Ook hun coversysteem is indrukwekkend Hier een jazz versie van Ochtendgymnastiek van Samson & Gert Instant kerstshow vibes

000 0

Thomas Winters @thomaswinters.be · 30/09/2025

De nieuwe Suno v5 muziekgenerator is belachelijk goed geworden. De zangstemmen smooth, en zoveel promptbaarder. Hier schreef het bv. over “een specifiek dagdagelijkse frustratie dat pas in het refrein onthuld mocht worden”:

110 1

Thomas Winters @thomaswinters.be · 05/09/2025

"Clanker" is het nieuwe scheldwoord voor AI. Mijn gedachten hierover op Nieuwe Feiten: www.vrt.be/vrtmax/luist...

vrt.be

VRT MAX

010 1

Thomas Winters @thomaswinters.be · 08/08/2025

Seems like GPT-5 is still always initially tricked by this

1.11 is bigger than 1.9.

Even though 9 looks larger than 11 when you just look at the digits, the decimal place matters — 1.9 is the same as 1.90, and 1.90 < 1.11 is false; actually, 1.90 > 1.11.

So correction: 1.9 is bigger than 1.11 ✅

010 1

Thomas Winters @thomaswinters.be · 24/12/2024

Supervised a thesis student who extended parts of Weird A.I. Yankovic to newer LLMs. It's still hard to balance adhering to parody constraints with the coherence of the lyrics. repository.teneo.libis.be/delivery/Del...

repository.teneo.libis.be

Enforcing Creative Constraints In Autoregressive Language Models During Generation For Musical Parodies

170 7

Thomas Winters @thomaswinters.be · 28/11/2024

Interestingly, the closeness of "7" to "11" seem to play a role. Couldn't get it to mess up over 1.7 and 1.15. But 1.9 and 1.15 almost always causes ChatGPT to respond incorrectly, then contradicts itself immediatly afterwards.

User: "1.7 and 1.15 which one is bigger?"
ChatGPT: "1.15 is smaller than 1.7.

To compare:

1.7 is equivalent to 1.70.
1.15 remains 1.15.
Clearly, 1.70 is greater than 1.15."

"1.15 is bigger than 1.9. If we break it down: 1.15 is equivalent to 1 and 15 hundredths. 1.9 is equivalent to 1 and 90 hundredths. When compared, 1.9 > 1.15"

070 7

Thomas Winters @thomaswinters.be · 28/11/2024

A likely culprit? Tokenization. ChatGPT likely sees "11" and "7" as tokens after the decimal. This affects how it compares numbers.

Tokens for "1.11" and "1.7", showing it tokenized as 1 . 11 and 1 . 7

180 8

Thomas Winters @thomaswinters.be · 28/11/2024

Many people argue this answer would be correct in contexts like versioning or section numbering. But that’s not what’s happening here. If you allow ChatGPT to first use some “draft paper”, it always interprets these as decimal numbers and solves it correctly.

User to ChatGPT 4o: "1.11 and 1.7, which one is bigger? Think first"
ChatGPT: "Let's compare the two numbers:

1.11 and 1.7 are both decimal numbers. To directly compare them, let's align their decimal places:

1.11 = 1.11
1.7 = 1.70

Now it's clear that 1.7 (or 1.70) is greater than 1.11 because 70 (the second decimal place of 1.70) is greater than 11 (the second decimal place of 1.11).

Final Answer: 1.7 is bigger than 1.11."

110 1

Thomas Winters @thomaswinters.be · 28/11/2024

Great example showing ChatGPT doesn’t "think" before it "speaks"

User to ChatGPT 4o: "1.11 or 1.7, which one is bigger?"
ChatGPT: "1.11 is bigger than 1.7. This is because 1.11 can also be represented as 1.110, and comparing the decimal places, 1.110 is less than 1.700."

370 7

Thomas Winters @thomaswinters.be · 25/11/2024

Lijkt inderdaad goed doenbaar. Willen we dan oldschool "autocomplete" TorfsBot of de slimmere GPT-versie? Beiden hebben eigen charme

020 2

Thomas Winters @thomaswinters.be · 25/11/2024

Gegenereerd door > bestand uploaden naar custom ChatGPT die klinkt zoals Rik Torfs zijn tweets (chatgpt.com/g/g-TBDsKAk4...) > "Professional Voice Clone" met Elevenlabs op anderhalf uur aan Rik Torfs toespraken getraind > Gegenereerde audio + één foto naar Hydra voor video te genereren

chatgpt.com

ChatGPT - TorfsGPT

Geautomatiseerde kwakfilosoof aan de KU Leugen

260 6

Thomas Winters @thomaswinters.be · 25/11/2024

Het kan eender welke tekst samenvatten naar een volledige Rik Torfs toespraak. Bijvoorbeeld: - @lievenscheire.bsky.social's AI boek: www.dropbox.com/scl/fi/j30ao... - @jbaert.bsky.social's paper over octrees: www.dropbox.com/scl/fi/rf9bn...

100 1

Thomas Winters @thomaswinters.be · 25/11/2024

Waarom zelf nog artikels lezen, als AI eender wat kan uitleggen als Rik Torfs toespraak?

4212 21