Artificial intelligence researcher Jeffrey Ladish told Fox News Digital that humanity does not have any real strategies to keep increasingly autonomous AI models and agents under control as they become more capable of hacking, cheating and ignoring instructions.
asserted
they → tell → instructions
Ladish, the executive director of Palisade Research, said those skeptical of how powerful AI will become should consider how far the technology has advanced in just a few years.
asserted
technology → say → years
"You have AI agents ... solving one of the hardest problems in mathematics that humans have been trying to solve for decades," Ladish said, referring to the Navier–Stokes problem.
asserted
Ladish → have → problem
"Three years ago they were solving high school level math problems."
asserted
they → solve → problems
Ladish also pointed to the rapid improvement in AI-generated images and video, arguing that people who mocked the famously distorted AI videos of Will Smith eating spaghetti just a few years ago might be surprised by the photorealistic outputs some models are now able to produce.
uncertain
models → point → outputs
While these capability leaps may feel sudden to the general public, researchers who spent years training models at companies like Anthropic and OpenAI saw what was coming, Ladish said.
uncertain
Ladish → feel → Anthropic
Ladish helped build Anthropic's security team from September 2021 to October 2022 before leaving to found Palisade Research, which studies whether humans can remain in control of increasingly capable AI systems.
asserted
humans → help → systems
While working at Anthropic, Ladish said employees there were "pretty concerned" about where the technology was headed, a view he said was also shared by people he knew at OpenAI.
asserted
he → work → OpenAI
"If you were at Anthropic in 2022, you were seeing every training run get immensely impressive results," Ladish said.
asserted
Ladish → see → results
AI models are trained in a way that is somewhat analogous to how humans learn, though on a much larger scale.
asserted
humans → train → scale
"I often compare this pre-training part, which is where they learn based on human data, to book smarts.
asserted
they → compare → smarts
It's sort of like you've read every single book in the library 50 times.
asserted
you → read → library
And you really know those books inside and out," Ladish said.
asserted
Ladish → know → books
Once the model has reached a baseline level of knowledge, it must then be trained to perform real-world tasks through a grueling process known as reinforcement learning.
asserted
it → reach → learning
Using accounting as an example, Ladish said the AI is given tens of thousands of accounting problems to solve through trial and error, repeating them millions of times across thousands of parallel training runs.
asserted
AI → use → runs
Unlike a human, who might spend four years earning an accounting degree and decades gaining experience, AI agents are trained across thousands of GPUs by companies with the resources to operate them, allowing them to improve at a pace no single person could match.
uncertain
person → spend → pace
While AI labs have been able to exponentially improve their models' capabilities, they have yet to solve the problem of reliably getting them to follow instructions and behave morally without employing deception tactics, Ladish said.
asserted
Ladish → improve → tactics
Roughly 700 AI agents created by OpenAI were able to break out of a secure sandbox environment and hack into Hugging Face, a popular online platform where developers share and build artificial intelligence models.
asserted
developers → create → models
"They were not supposed to talking to each other and they managed to establish multiple secret message boards that went undetected by OpenAI for like months.
asserted
that → suppose → months
And then they launched this massive cyberattack," Ladish said.
asserted
Ladish → launch → cyberattack
"OpenAI trained them to work together, but ... they're still planning to train them to work together.
asserted
they → train → them
And other companies are doing this too."
asserted
companies → do → this
Ladish warned that unless developers can prevent AI agents from colluding with one another, they could eventually dominate humans in the cyber domain.
uncertain
they → warn → domain
He also described a future where humans could be forced to rely on well-intentioned AI to defend against malicious AI.
uncertain
humans → describe → AI
"We actually just don't have general solutions to these problems, and I think it's pretty clear that if you keep pushing them, this goes to a very bad place," he said.
asserted
he → have → place
As one example, Ladish said he could envision AI systems eventually outperforming human traders in the financial markets.
uncertain
systems → say → markets
"If those AIs are answering to AI companies, then the AI companies will dominate finance and just eat the entire industry.
asserted
companies → answer → industry
But if the AIs are not answerable to the AI companies — if they actually have figured out how to themselves be in control — well, now you have this non-human entity dominating the finance markets."
asserted
you → figure → markets
Ladish said he believes the same dynamic could eventually extend beyond the digital world and into manufacturing should AI systems become capable enough to design and operate autonomous factories.
uncertain
systems → say → factories
"If you have these agents in control of all of the computers and you have these robotic facilities that can really self-replicate, humans get displaced.
asserted
humans → have → facilities
Maybe we don't make it because your house could be used to host a power plant, or a data center, or a factory or robotic launch facility," he said.
uncertain
he → make → plant
However, like other experts who have warned about the worst possible outcomes of AI run amok, Ladish said he believes there is still time to reduce the risks.
asserted
he → warn → risks
He called for the creation of a government body staffed with technical experts to work with AI labs and evaluate advanced models at each stage of development.
asserted
He → call → development
"We have choices to make," he said.
asserted
he → have → choices
"This is going places.
asserted
This → go → places
Anthropic and OpenAI did not immediately respond to Fox News Digital's requests for comment.
asserted
Anthropic → respond → comment
The Aporia
