Sounds like Newton. Long tradition, I guess

"Okay, nuclear power. Yeah. Yeah, they did give us that. That's true. "

"And industrial-scale superconducting magnets!"

"Don't forget GPS satellite relativistic corrections!"

"And radioisotope therapy!"

"Electron beam therapy?"

"World Wide Web, wasn't that created by particle physicists?"

"Yeah, they did do that."

"Capacitative touchscreens too, you know."

"All right, all right, FAIR ENOUGH! But APART from computers, telecommunications, nuclear power, PET scans, MRIs, satellite communications, superconductors, radioisotope therapy, electron beam therapy, heat-shrink tubing, cargo and baggage screening, smoke detectors, muon tomography, capacitative touchscreens, and the World Wide Web, WHAT HAS PARTICLE PHYSICS EVER DONE FOR US?!"

"Synchrotron light sources for drug discovery?"

"Oh, drug discovery?! GROW UP!"

Yeah, what has a better understanding of particle physics ever done for US, huh??? ... ...

_ ...........

if this were true it would not be possible for the system to read the watermark.

You confuse "invisible to humans" with "invisible to statistical analysis". Humans can't, for example, see the difference between metameric colours. Statistical analysis can. Humans can't see minor drift at scales where Brownian motion dominates. Statistical analysis can. Humans can't see the difference between background disease rates and new rising threats. Statistical analysis can. On and on. Statistical analysis is a far more powerful tool than human perceptual abilities.

this is also a lie. Identical thoughts must be expressed identically, otherwise they are, in fact, *different* thoughts.

I literally gave an example above. If you saw "picked a mango", "selected and picked a mango", "from within the tree, picked a mango", you would never think there was anything going on in the choice, and it isn't going to change the story one iota.

("It actually makes RAM needs worse" -> "MoEs actually make RAM needs worse")

It actually makes RAM needs worse than a dense model, for a given quality.

That said, inference techniques like FreeToken are improving MoE swapping performance, so it's not as bad as it once was. Honestly, I would not be at all surprised if we start doing training in a swap-aware manner, where expert cache misses count as loss and models can queue experts to start loading before they're needed. Could even train in two tiers of loading - RAM and disk. Wouldn't that be great? Could run multi-terabyte models on consumer hardware. Training would tend to tend to concentrate key logic and reasoning on small number of very active experts, mixed reasoning/knowledge on less common experts that are usually kept in RAM, and rarer knowledge on experts that usually remain on-disk until needed.

There are well known, reputable abliterators on HuggingFace.

Also, censorship usually (not always) is something you care about for chats, not agentic work.

A .safetensors or .gguf model cannot run code unless you let it. No agentic framework = all it can do is write text. It can't hurt you directly. At worst, it can only lie to you.

It's big, but not good for its size. They're using tricks to pretend that they're better than they are. For example, compare the numbers that they list for the competion on DeepSWE up against the actual DeepSWE scores.

Not cool, Mistral.

Artificial Analysis is overrated, but yes, Mistral is playing fast and loose with their claims.

With permanent DST we won't get the stories twice a year.

It takes one hell of a toll to try to do verbal exams 1:1 on students to evaluate them.

Nobody said "verbal exams 1:1".

The only thing that is required is that they be separated from their computers and their phones.

Nobody, I repeat, nobody who uses AI is going to see "OpenAI adding watermarks" and think, "Oh, I better use AI more, especially OpenAI".

And nobody, I repeat, nobody who doesn't use AI is going to see "OpenAI adding watermarks" and think, "Oh, I better use AI more, now that it'll be easier to see that what I did is made by AI."

It is not "an advertisement of the power of the technology". It's a regulation forced on them by the EU's AI Act, which neither companies nor users want.

Perhaps you should be realizing by now that homework is no longer a viable way to test how well students actually know the subject matter, and the only viable way to do so is in-class tests/exercises - frequent enough and worth enough of the grade to force them to study between lessons.

OR, alternatively, you could just keep doing what you're doing, and instead, in effect grade people on:

Comparably poor grades:
* People who actually do the work by hand
* "AI N00bs" who let stuff like that slip through

Comparably good grades:
* AI-experienced people who know enough to skim over the work, do multiple passes, or issue prompts that prevent such tells from getting into the work

Wherein, in effect, your net reward/punishment is inverted vs. what you should want it to be.

Closing one's eyes and ears to the fact that the environment students operate in has changed isn't an acceptable option. Education must adapt. And watermarks, BTW, will only catch "the n00bs"; more experienced / up to date students will keep track of what watermarks and what doesn't, and use AIs that don't watermark, or de-watermarking services.

I'll repeat: watermarks are anti-marketing. They discourage use.

SynthID in text genuinely is invisible. The probability shifts are small, and all the pathways are valid pathways to express the same thought.

And no, you cannot "collide" text watermarks. At best, it'll only have the latter one's watermark. At worst, both.