Sam Altman Responds To ‘incendiary’ New Yorker Article After Attack On His Home

OpenAI CEO Sam Altman published a blog post connected Friday evening responding to some an evident onslaught connected his location and an in-depth New Yorker profile raising questions astir his trustworthiness.

Early Friday morning, personification allegedly threw a Molotov cocktail astatine Altman’s home. No 1 was wounded successful nan incident, and a fishy was later arrested astatine OpenAI headquarters, wherever he was threatening to pain down nan building, according to nan San Francisco Police Department.

While nan constabulary person not identified nan fishy publicly, Altman noted that nan incident came a fewer days aft “an incendiary article” was published astir him. He said personification had suggested that nan article’s publication “at a clip of awesome worry astir AI” could make things “more dangerous” for him.

“I brushed it aside,” Altman said. “Now I americium awake successful nan mediate of nan nighttime and pissed, and reasoning that I person underestimated nan powerfulness of words and narratives.”

The article successful mobility was a lengthy investigative portion written by Ronan Farrow (who won a Pulitzer for his reporting that revealed galore of nan intersexual maltreatment allegations astir Harvey Weinstein) and Andrew Marantz (who’s written extensively astir exertion and politics).

Farrow and Marantz said that during interviews pinch much than 100 group who person knowledge of Altman’s business conduct, astir described Altman arsenic personification pinch “a relentless will to powerfulness that, moreover among industrialists who put their names connected spaceships, sets him apart.” 

Echoing other journalists who person profiled Altman, Farrow and Marantz suggested that galore sources raised questions astir his trustworthiness, pinch 1 anonymous committee personnel saying he combines “a beardown desire to please people, to beryllium liked successful immoderate fixed interaction” pinch “a sociopathic deficiency of interest for nan consequences that whitethorn travel from deceiving someone.”

In his response, Altman said that looking back, he tin place “a batch of things I’m proud of and a bunch of mistakes.”

Among nan mistakes, he said, is simply a inclination towards “being conflict-averse,” which he said has “caused awesome symptom for maine and OpenAI.”

“I americium not proud of handling myself severely successful a conflict pinch our erstwhile committee that led to a immense messiness for nan company,” Altman said, presumably referring to his removal and accelerated reinstatement arsenic OpenAI CEO backmost successful 2023. “I person made galore different mistakes passim nan insane trajectory of OpenAI; I americium a flawed personification successful nan halfway of an exceptionally analyzable situation, trying to get a small amended each year, ever moving for nan mission.”

He added, “I americium sorry to group I’ve wounded and wish I had learned much faster.”

Altman besides acknowledged that location seems to beryllium “so overmuch Shakespearean play betwixt nan companies successful our field,” which he attributed to a “‘ring of power’ dynamic” that “makes group do crazy things.”

Of course, nan correct measurement to woody pinch nan ring of power is to destruct it, truthful Altman added, “I don’t mean that [artificial wide intelligence] is nan ringing itself, but alternatively nan totalizing accuracy of ‘being nan 1 to power AGI.’” His projected solution is “to orient towards sharing nan exertion pinch group broadly, and for nary 1 to person nan ring.”

Altman concluded by saying that he welcomes “good-faith disapproval and debate,” while reiterating his belief that “technological advancement tin make nan early unbelievably good, for your family and mine.”

“While we person that debate, we should de-escalate nan rhetoric and strategies and effort to person less explosions successful less homes, figuratively and literally,” he said.