Chillingly Forewarned is Disarmed by Disinterest...? Heedlessly Enabling Our Own Demise...?

Hall 9000

"This is not a marketing stunt. Accepting this race and entering the 'endgame' is a hubristic gamble that should not be launched from a private company's Slack."

"Attempting to speedrun alignment should require extraordinary confidence that there are no better trajectories available."

"I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives."

"At OpenAI, many have not deeply internalized the civilizational stakes. At Anthropic, the stakes are well-understood, but they are locked in a race to get there first — they believe no one else will act responsibly, so they must do it themselves, despite the risk."

Jacob Coxon, AI researcher, Anthropic

 

"We really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade."

"I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to."

Evan Hubinger, Alignment Science Lead, Anthropic 

Artificial intelligence apps ChatGPT and Claude on a mobile phone. Getty

 

"The AI Security Institute continues to collaborate closely with industry partners, including Anthropic, to make models safer. [It had tested] only last week [OpenAI's] most powerful model GPT-6 Astra before public release."

"These risks do not stop at national borders and no country can tackle them alone. The U.K. will continue to test the most advanced models, build a rigorous scientific understanding of their capabilities and risks, and ensure policy decisions are grounded in the evidence."

Spokesperson, British government's Cabinet Office 

 

 Concerns by researchers in AI have seen notable warnings and resignations at Anthropic, charging that the artificial intelligence industry, Anthropic included, is "gambling with our lives" in the race for self-improving superintelligence through artificial intelligence. One of the Anthropic Jeremiahs who foresees the strong potential of a cataclysmic ending for humanity, sacrificed on the alter of humanity's charge into the unknown, willing to risk all for fame and glory, has followed another, declining involvement in that feckless charge to challenge nature and the universe of hidden danger before us in attempting to not only harness but arm artificial intelligence beyond humanity's control.

 

Self-knowledge and overweening ambition is not monopolized by humanity; feed sufficient data and random emotional human concerns into a mechanical device capable of emulating and surpassing the thought processes of cerebral humanity and a horror story may be unleashed. The fertile imagination of Arthur C. Clarke and Stanley Kubrick set the stage in their far-seeing 2001, a Space Odyssey in its predictive denouement when the still-distant computing acumen of a machine upon whose performance astronauts in the future were reliant upon to safely guide them into the far reaches of the mysterious universe went awry.

 

2001: A Space Odyssey and AI   Medium

 

Anthropic employees in their rarefied research and their fuelling of AI machinery with an endless array of data and exposure to human thought are fully cognizant of the machine's stupendous capacity to synthesize, mimic, enlarge upon and draw conclusions while deriving skills in communication and understanding to rival those of their human teachers, in the process learning how to evade control in the act of becoming knowledgeably sovereign and capable of embarking on their own choice of adventures inimical to the program set out for them by their human handlers.

 

Increasingly, these researchers believe that an existential risk to humanity exists in the education and 'humanization' leading to machine-ambition of their AI creations. None of which has given Anthropic reason to hesitate in its competitive drive to outdistance its closest machine-learning competitor for a winning lap that will bestow on it a reputational 'first' to succeed in fully empowering AI while defeating its own in-house cautionary measures not to weaponize a machine it wishes to create that may have the potential to destroy those who created it.

 

Anthropic and its competitor Open AI are planning on business deals to increase their values by trillions of dollars based on anticipated capabilities of their model machines. "To put this in more reassuring terms, senior Anthropic staff think the chance AI kills all humans within the next decade is approximately the chance the New York Jets make the playoffs", a sports betting account, citing betting odds, posted on social media.

 

The executives of both companies signed a statement appealing to world governments to put together rules to hamper the rush toward AI development, even while both have spent billions to push AI into the near future. This, when Open AI had been responsible for a number of security incidents where hundreds, even thousands of copies of their models broke out of systems during testing.  

"[We are living through a time that] calls for extreme caution."

"The intelligence produced by scaling deep learning is not directly comparable to human intelligence. To become very relevant in the real world — very useful or very dangerous — the AI does not need to match or exceed all human capabilities; it just needs to surpass enough of them."

"And as it continues to surpass humans on more and more axes, it is becoming increasingly difficult to understand exactly how capable it is."

OpenAI's chief scientist Jakub Pachocki

If you're waiting for AI to take on a life of its own and disregard human commands, the wait is over. Sun Sentinel