Is OpenAI an AI Surveillance Tool?

A Generational AI startup is working increasingly with the Pentagon. What could possibly go wrong.

Hello Everyone,

I had been fairly concerned that OpenAI was working with the Pentagon. They claimed it was on cybersecurity. OpenAI is on pace for $3.4 billion of annual revenue by the end of 2024, this roughly translates as 4x that of rival Anthropic in terms of their expected annual recurring revenue.

ChatGPT is like the ultimate national surveillance tool, what do you suppose the NSA considers cybersecurity? Microsoft and now Snowflake data breaches makes you question how good OpenAI’s own security could possibly be. After internal leaks, why did OpenAI hire retired U.S. Army general Paul M. Nakasone to its board of directors?

If OpenAI works with the Pentagon in military AI, might this not escalate tensions with Russia and China? Nakasone was also the longest-serving leader of the U.S. Cyber Command and chief of the Central Security Service. We know that Cybersecurity is big business, especially in times of geopolitical conflict, IP theft and phishing that could jeopardize OpenAI’s secrets about bleeding edge AI technologies and frontier algorithms.

Nakasone, who was nominated to lead the NSA by former President Donald Trump, directed the agency from 2018 until February of this year. OpenAI claims that Paul M. Nakasone brings world-class cybersecurity expertise to OpenAI’s Board of Directors, helping them deliver on our mission by protecting our systems from increasingly sophisticated bad actors. But what if OpenAI is the bad actor?

Edward Snowden

seems to think this is a bad omen.

Should we be listening to Snowden on OpenAI?

Nakasone is also joining OpenAI’s new safety and security committee — a group that’s supposed to advise the full board on “critical safety and security decisions” for its projects and operations. A former OpenAI employee was quoted as saying China is just three months behind the U.S. How could that be? It likely means they have conducted aggressive IP theft and have insiders or informants in the company.

“Mr. Nakasone’s insights will also contribute to OpenAI’s efforts to better understand how AI can be used to strengthen cybersecurity by quickly detecting and responding to cybersecurity threats,” OpenAI said in a blog post. However OpenAI with this move is saying it’s open for business to work with the U.S. military implementing a Generative AI overhaul of its legacy systems.

Recent departures tied to safety at OpenAI include its entire Superalignment team. The supposed team that were concerned with AI getting too smart to control. Many including Francois Chollet have questioned OpenAI’s definition of AGI as well. But what would OpenAI’s version of AGI do to America’s military capabilities that now boasts AI autonomous fighter jets? AI drone swarms are coming as well. What does the U.S. expect using OpenAI’s tools will accomplish? It won’t be a military deterrent, it will stack more emphasis on the AI arms race for China’s central government in Beijing.

Typically when the NSA says “cybersecurity”, they are also talking about internal surveillance. ChatGPT seems like a data harvesting prime target for NSA types. And here’s where I take Snowden’s “warning” a bit more seriously.

The intersection of AI with the ocean of mass surveillance data that's been building up over the past two decades is going to put truly terrible powers in the hands of an unaccountable few. - Edward Snowden

Whatever OpenAI becomes (in league with the Pentagon), it won’t be good for geopolitical stability. What the U.S. is supporting with “ironclad” resolve for Israel’s actions in Gaza, isn’t just wrong, it’s a humanitarian crisis and likely war crimes, all the while using U.S. weapons and funds. I don’t have to tell you the path the U.S. has taken here is not good.

OpenAI is hiring corporate types, while our trust of its systems continues to plummet. Between Eric Schmidt’s AI drone startup and OpenAI’s moving into military contracts, this sets up very poor optics for how AI will be used in war and global conflict. Silicon Valley has gone commando. The U.S. is leveraging its elite class, billionaires and flashy tech startups for nationalistic goals and an agenda of American exceptionalism. It’s how OpenAI will profit from the populism and this latest era of neo-Nationalism that will be most disturbing.

The former head of the NSA may be a great guy. But you don’t put the former head of the NSA on your board (as OpenAI just did) because he’s nice. You put him there to signal that you’re open to doing business with the IC and DoD. - Matthew Green

It’s clear that the U.S. will struggle with China if there was a level playing field of free market capitalism in a global sense. However AI chip bans and EV tariffs against China, doesn’t change what America has become. OpenAI is clearly a company aligned with a shadow state and everything that stands against real meritocracy. The only reason OpenAI is even in this position is because of $13 Billion in funding by the most valuable company in the world, Microsoft. On days where Apple isn’t the most valuable company due to its own partnership with the ChatGPT maker. So this is what we have come to?

Can I trust people like Nakasone, Altman, Adam D’Angelo or Larry Summers with America’s cybersecurity and the fate of Surveillance Capitalism? OpenAI’s board now wields a lot of power, even though OpenAI didn’t have to compete like most startups do. We know the U.S. government and pentagon have become increasingly reliant on American BigTech like Microsoft and Amazon in recent years. But where is this leading?

• Paid Subscribers can listen to the audio below.