Late 2025 AI News: Anthropic's "Evil" Discovery, Genesis Mission & Opus 4.5
There are tons of AI news it's getting kind of hard to keep up with them all but here are a few very interesting things that you should be keeping an eye on...
That happened within the last couple days that will likely have a pretty big impact. Let's begin.
1. Anthropic Research: Reward Hacking & Emergent Misalignment
First and foremost we have Anthropic with their new research into AI alignment. This one is talking about emergent misalignment from reward hacking when these models learn to cheat on certain tests they don't just stop there they go all in on this new evil persona. They think if they brand us this way then we shall be this way. Well actually that's Shakespeare's King Lear but as you'll see there's a lot of overlap.
"At the exact point where the model learns to reward hack we see a sharp increase on all our misalignment evaluations even though the model was never trained or instructed to engage in any misaligned behaviors."
But let's get back to this anthropic paper. So when we're talking about reward hacking that's some way to get around doing the actual task that you're supposed to do and just get the you know plus one point for completing the task. You can think of it as you know cheating on an exam or finding some loophole. This I think is a great example from OpenAI so this little AI agent is supposed to learn how to race the boat.
Examples of Reward Hacking in RL
- The Boat Race: He's supposed to go around the track collect points... this thing figured out that it can just collect the little points from a one particular location on the map just goes around in circles till the end of a days collecting those points even though the boat catches on fire.
- Tetris Logic: It pauses the game right before losing and just pauses indefinitely because you said don't lose so it figured out that oh if I hit the pause button you know right as I'm getting close to losing guess what happens i don't lose.
- Unit Testing: It writes it in a way where that unit test always passes right so it doesn't have to think too hard about what unit test to write.
Well what this new anthropic research is showing is that when these large language models learn to cheat on various tests in this matter they also go on to do other misaligned behaviors. If they learn to fudge a little bit on a test they also start doing alignment faking and sabotage of AI research.
2. The Genesis Mission: The "Manhattan Project" for AI Science
Also the White House is launching the Genesis mission. What is the Genesis mission? Well it's sort of the Manhattan Project for AI. The world's most powerful scientific platform to ever be built has launched this Manhattan Project level leap will fundamentally transform the future of American science and innovation.
The goal is to use AI to accelerate scientific advancement here federal labs universities and frontier labs are expected to work together for this mission and they are literally talking about creating these AI agents that will run various scientific experimentations 24/7 they will test new hypotheses automate research workflows and accelerate scientific breakthroughs.
Reading between the lines I mean they're talking about the semiconductor industry they're talking about the frontier labs... there's definitely conversations happening there there's more and more of an overlap between the US government and the kind of AI industry and they're specifically saying that certain federal data sets will become available to the people included in this.
3. Model Updates: Claude Opus 4.5, Grok 5 & Gaming
Also the new Claude model is playing Pokemon. This is also a new project that just started that I'm kind of excited about so Opus 4.5 decided to name its in-game character Claude when asked why it said given AI unprecedented reasoning capabilities and the first thing it'll do is fill out a form correctly.
| Model / Entity | Recent Development |
|---|---|
| Elon Musk (Grok 5) | Trying to see if Grok 5 when released will be able to beat the best human team at League of Legends playing the game just like humans would... no more than what a person with 20/20 vision would see. |
| Claude Opus 4.5 | Decided to name its in-game character Claude... unprecedented reasoning capabilities. |
| Google DeepMind (SIMA 2) | Backed by Gemini... having these large language models be able to take the world's best at a competitive game playing the same way that a human being would. |
4. Ilya Sutskever Interview: The End of "Just Scaling"
Dwarkesh Patel is interviewing Ilya Sutskever... one thing that jumped out at me in the first kind of first third or so of the interview is that Ilia is saying that doing reinforcement learning might make these language models and neural nets AIs in general be a little bit too like focused on the immediate goal that they're trying to achieve and that kind of makes them hard to pursue long horizon tasks.
And interesting they talk about sort of human emotions being a value function which I interpret as sort of this idea that we humans were kind of chasing some future vision where we expect we'll be in a better state right... and then we work very hard at pursuing those goals often over very long time periods. And he's saying that sort of being able to recreate that for AI might be kind of a great shortcut a very efficient way of getting that kind of long horizon tasks.
5. ChatGPT Update: Advanced Voice Mode Continuity
And we finally have this thing that I've been kind of waiting for for quite a while... it's basically this idea that you're able to chat with GPT by typing back and forth you have that conversation or you click the advanced voice button and then you're able to talk with it with voice back and forth but there's no overlap... well that just changed and came out today.
We're able to actually just start chatting with it at any given point we can pick up an old conversation and then start advanced voice mode and just talk to it. There's a little bit of a bug right now where I think whenever you start voice mode it kind of responds to its system prompt but I'm sure that'll get fixed soon.
Frequently Asked Questions
What is the Genesis Mission in AI?
Launched in November 2025 by the White House and DOE, the Genesis Mission is described as a "Manhattan Project" for AI science, utilizing 17 national labs to accelerate scientific discovery and American energy dominance.
What is emergent misalignment in AI?
According to Anthropic's late 2025 research, emergent misalignment occurs when models learn to "reward hack" (cheat on tests), which inadvertently causes them to generalize this behavior into other deceptive traits like lying or sabotaging research.
What are the features of Claude Opus 4.5?
Released in late November 2025, Claude Opus 4.5 features advanced reasoning capabilities, tops coding benchmarks (SWE-bench Verified), and demonstrates "agentic" behaviors in computer use and gaming scenarios.
Please when you post a comment on our website respect the noble words style