I was very lucky to be involved in Open Ocean training with Dolphin Quest during my internship there and I’m so happy to see it evolve into what it is today.

Cirrus is a former navy dolphin, so she took to the open ocean very easily and is the one who does the open ocean walks.

When I was there, there was still the netted ocean habitats. The net was there to provide the dolphins with extra security when going out there. But damn that thing was a pain to maintain. That thing needed to be inspected manually on any day ocean habitat training was happening - ensuring there was no damage to it.

Then it had to be cleaned every 6 months and it took a team of interns and trainers two whole days to remove it, clean it and put it back in. That was a loooong two days.

For Ocean Habitat training we were still working on encouraging some of the younger dolphins like Cooper, Marley and Cavello to go out to a netted habitat.

The dolphins were very hesitant and sometimes would bolt back through the tunnel if they felt it was too much. Other dolphins just flat out refused to enter the tunnel entrance lagoon because even that was too much for them.

A lot of people have this idea that dolphins in human care are desperate for more space and would swim off without a second thought. The reality is that the dolphins don’t understand why they should leave the comfort and safety of what they know to be their home and venture into unknown environments. How are they supposed to know that’s actually ~natural~ and ~better~ for them? They don’t care about that, they care about what is reinforcing, what feels safe and where their social groups are.

We literally have to convince them and reinforce them and show them it’s okay and trainers really need to have a good relationship with their animals for this to happen.

Honestly I was so surprised in seeing these reactions and seeing how the dolphins responded to being offered larger habitats. They really just. Didn’t care.

We created a reinforcement history around the open ocean and the ocean habitat. It didn’t offer any intrinsic value and the dolphins only showed interest and motivation in going out there when they were reinforced for it.

It was noted by trainers that the males were usually braver and easier to train for open ocean work than the females, and some speculated this was due to the males only having themselves to worry about, while the females also kept their calves and pod in mind.

I had the wonderful chance to accompany the dolphins outside a few times, once with a mum and son, the other times with a male duo. More than anything they seemed to enjoy the chance to spend time playing together in the waves and with their trainers. More than they ever paid attention to being "home" in the wide open ocean.

This ukiyo-e was painted in 1934 by the famous Itô Shinsui 伊東 深水 (1898-1972).

This lady is performing theLion Danceliterally named 鏡獅子 Kagami jishi (Lion Dance) 1934.

“Ghosts aren’t real” it’s very fair and rational that you believe that

“Ghosts aren’t real anymore” I’m about to hear a poem or very sad story

This is the quiz and I did not come up with it, credit goes to @iswearbyalltheflowers on this post.

Not sometimes. Not until the next update. Always. They proved it with math.

Even with perfect training data and unlimited computing power, AI models will still confidently tell you things that are completely false. This isn't a bug they're working on. It's baked into how these systems work at a fundamental level.

And their own numbers are brutal. OpenAI's o1 reasoning model hallucinates 16% of the time. Their newer o3 model? 33%. Their newest o4-mini? 48%. Nearly half of what their most recent model tells you could be fabricated. The "smarter" models are actually getting worse at telling the truth.

Here's why it can't be fixed. Language models work by predicting the next word based on probability. When they hit something uncertain, they don't pause. They don't flag it. They guess. And they guess with complete confidence, because that's exactly what they were trained to do.

The researchers looked at the 10 biggest AI benchmarks used to measure how good these models are. 9 out of 10 give the same score for saying "I don't know" as for giving a completely wrong answer: zero points. The entire testing system literally punishes honesty and rewards guessing.

So the AI learned the optimal strategy: always guess. Never admit uncertainty. Sound confident even when you're making it up.

OpenAI's proposed fix? Have ChatGPT say "I don't know" when it's unsure. Their own math shows this would mean roughly 30% of your questions get no answer. Imagine asking ChatGPT something three times out of ten and getting "I'm not confident enough to respond." Users would leave overnight. So the fix exists, but it would kill the product.

This isn't just OpenAI's problem. DeepMind and Tsinghua University independently reached the same conclusion. Three of the world's top AI labs, working separately, all agree: this is permanent.

Got the link via @finalducc

The governments agenda is clearly spelled out in this consultation, through the phrasing of the questions.

Excuse me while I link a video. It’s a clip from a comedy so it’s a bit exaggerated but it demonstrates the point very well.

Almost all the questions in that survey were “leading questions”. They had an inherent presumption that the person answering agreed with age restrictions, wanted more severe ones and wanted more websites and services restricted. (Believe me I take no pleasure in being correct that VPN’s are next on their target list).

This is a consultation where they try and make you look a monster for saying no. They push the “but it’s for the children” hard. Like there was one question where in order to say no, you had to not object to children sending/receiving nudes which felt wrong. I mean that’s bad, obviously that’s bad. But the answer isn’t total lockdown of the internet to verified ID.

So yeah if you are from the UK then fill this out, and fill it out carefully. Don’t let them manipulate or trick you into agreeing that mandated age restrictions are necessary, as that is what we have with this “you must prove your identity” and it looks like they want to make it worse, not roll it back.