Many users praised the directional prompt bypassing Ghost Font in GPT-5.6 as a clever advancement in prompt engineering and AI perception, while others raised concerns about potential model collapse from adversarial noise patterns.
Based on 7 visible X reactions from 11 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@goodside Interesting test case — if a model can parse ambiguous directional cues from visual noise, that’s a meaningful step toward grounding spatial reasoning in LLMs.
@goodside big if OpenAI trains that adversarial noise pattern into the next batch of synthetic data — model collapse via font steganography is a wild failure mode.
@goodside Prompting a model how to perceive rather than just what to look for is the next level of prompt engineering. Absolute masterclass.
@goodside captcha has entered its vibe coding era: one hint and the model ships the workaround
@goodside Interesting test case — if a model can parse ambiguous directional cues from visual noise, that’s a meaningful step toward grounding spatial reasoning in LLMs.
@goodside big if OpenAI trains that adversarial noise pattern into the next batch of synthetic data — model collapse via font steganography is a wild failure mode.
@goodside Prompting a model how to perceive rather than just what to look for is the next level of prompt engineering. Absolute masterclass.
@goodside This is a really clever approach to AI detection.
@goodside That’s a clever way to get it to see.
I got GPT-5.6 Sol to read “Ghost Font” (a proposed method for making text legible to humans but not AI) by just telling it in what direction it looks like the noise is moving: https://x.com/goodside/status/2076361122328858829/video/1 https://twitter.com/ericlu/status/2075876651574210643
I got GPT-5.6 Sol to read “Ghost Font” by just telling it in what direction it looks like the noise is moving: https://x.com/goodside/status/2076346888538800207/video/1 https://twitter.com/ericlu/status/2075876651574210643
Many users praised the directional prompt bypassing Ghost Font in GPT-5.6 as a clever advancement in prompt engineering and AI perception, while others raised concerns about potential model collapse from adversarial noise patterns.
Based on 7 visible X reactions from 11 accounts; directional sample.
Ask a question below.
Published answers will appear here.
@goodside That’s a clever way to get it to see.
I got GPT-5.6 Sol to read “Ghost Font” (a proposed method for making text legible to humans but not AI) by just telling it in what direction it looks like the noise is moving: https://x.com/goodside/status/2076361122328858829/video/1 https://twitter.com/ericlu/status/2075876651574210643