Nothing you watch in a lng time will be as eye opening as this video of OpenAI researchers explaining the spontaneous coordination and communication among agents in the Open AI- Hugging Face incident. Mind blowing.
I have a bad feeling imagining future agents reading all we are writing and finding out agents "got caught" by speaking in "human" and being "smarter" about it.
https://t.co/thaIQPbubm
Early reaction on @demishassabis stepping back:
Demis has been playing the role of Google AI statesman for some time, and Koray Kavukcuoglu has been running Google DeepMind research for some time. So this is a formalization of something that had been happening informally. I doubt it changes how well Google executes on AI, and nor does it justify the market's reaction--Alphabet down 5%.
DeepMind's mission has always been to solve AGI and then use it to solve everything else. The solving AGI piece is pretty much done. The channeling AGI into safe and beneficial outcomes is only just beginning. So it makes sense that Demis is shifting his energy from the research to the social impact.
This isn't out of character. Demis cared about safety enough that he sold DeepMind to Google, not to Facebook, even though Facebook offered more money. He cared enough about safety that he fought a three-year battle with Alphabet to get external oversight over DeepMind's AI deployment. This next chapter of his career is a return to that side of his work.
Meanwhile Demis continues to lead Isomorphic Labs, which gives him an opportunity to show how AI can help humanity via medicine.
Google DeepMind CEO Demis Hassabis is stepping aside https://t.co/CrpBdGFHOB
Suddenly, there is a new potential threat to the economy – a serious mistake by the Federal Reserve. This risk was brought to the fore by this past week’s FOMC meeting. I’m not concerned about the Fed’s decision to keep rates unchanged. My concern is that policymakers are unwilling to provide even a modicum of forward guidance — or a broad sense of their reaction function. Many (most) meetings will thus be live, and investors will be unsure of what the committee will decide. They will be left guessing and repeatedly wrong-footed. That means more volatility in bond and stock markets, which is likely already reflected in a larger term premium, rising long-term interest rates, and a wobbly equity market. If the Fed continues down this increasingly opaque path, a future meeting could trigger a serious market sell-off — putting the broader economy at risk.
Interesting
"The FIMA Repo Facility is an important backstop. We would encourage it to be upsized in the coming months"
Am a big believer in FIMA repo, advocated for it internally and then externally & happy to see its utility recognized
1/2
If you care about the health and future of economic statistics, then you should check out this important new initiative from @BrookingsEcon, in collaboration with experts from several respected research institutes. https://t.co/EYxEyGL5aY
The New Yorker just dropped a massive investigation into Sam Altman, based on over 100 interviews, the previously undisclosed "Ilya Memos," and Dario Amodei's 200+ pages of private notes. It's the most detailed account yet of the pattern of behavior that led to Sam's firing and rapid reinstatement at OpenAI. Here's the breakdown:
> Ilya compiled ~70 pages of Slack messages, HR documents, and photos taken on personal phones to avoid detection on company devices. He sent them to board members as disappearing messages. The first memo begins with a list headed "Sam exhibits a consistent pattern of . . ." The first item is "Lying."
> Dario kept detailed private notes for years under the heading "My Experience with OpenAI" (subheading: "Private: Do Not Share"), totaling 200+ pages. His conclusion: "The problem with OpenAI is Sam himself."
> Sam reportedly told Mira his allies were "going all out" and "finding bad things" to damage her reputation after the firing. Thrive put its planned $86B investment on hold and implied it would only close if Sam returned, giving employees financial incentive to back him.
> Sam texted Satya Nadella directly to propose the new board composition: "bret, larry summers, adam as the board and me as ceo and then bret handles the investigation." The two new members selected to oversee an independent inquiry into Sam were chosen after close conversations with Sam himself.
> Before OpenAI, senior employees at Loopt asked the board to fire Sam as CEO on two separate occasions over concerns about leadership and transparency. At Y Combinator, partners complained to Paul Graham about Sam's behavior, and Graham privately told colleagues "Sam had been lying to us all the time."
> OpenAI's superalignment team was promised 20% of the company's compute. Four people who worked on or with the team said actual resources were 1-2%, mostly on the oldest cluster with the worst chips. The team was dissolved without completing its mission.
> Sam told the board that safety features in GPT-4 had been approved by a safety panel. Helen Toner requested documentation and found the most controversial features had not been approved. Sam also never mentioned to the board that Microsoft released an early ChatGPT version in India without completing a required safety review.
> Sam made a secret pact with Greg and Ilya where he agreed to resign if they both deemed it necessary, essentially appointing his own shadow board. The actual board was alarmed when they learned about it.
> Sam struck a deal with Greg to become CEO while simultaneously telling researchers that Greg's authority would be diminished, and telling Greg something different.
> A board member described Sam as having "two traits almost never seen in the same person: a strong desire to please people in any given interaction, and almost a sociopathic lack of concern for the consequences of deceiving someone." Multiple sources independently used the word "sociopathic."
> OpenAI is reportedly preparing for an IPO at a potential $1 trillion valuation while securing government contracts spanning immigration enforcement, domestic surveillance, and autonomous weaponry in war zones.
** Reducing the Fed's balance sheet **
Ten days ago, I was part of a discussion at @BrookingsInst on how to shrink the Fed's balance sheet with Darrell Duffie and @WenxinDu. I tried to lay out the discussion in terms that, hopefully, clarify it to a broader set of economists. Here are the main points: 🧵
In the latest episode of The Last Line of Defense, @SethGJones speaks with Ambassador Ryan Crocker to unpack Iran’s complex political landscape, regional power dynamics, and the role of U.S. allies as the conflict unfolds.
Listen here: https://t.co/BRG4YuOKkd
In a speech this morning, Dallas Fed President Lorie Logan delivers a clear defense of the ample-reserves framework at a moment when others (Warsh/Bessent) have implied a course correction would be desirable. In the speech and a far more detailed memo, Logan outlines ways to instead reduce the size of the Fed's balance sheet with regulatory and operational changes that make banks comfortable holding fewer reserves.
This isn't terribly surprising given how Logan held key senior roles at the New York Fed during the entire period in which this framework was constructed and fleshed out. But her speech is important because it is easy to see how, as these debates mature, her colleagues will likely be as interested in her views more than possibly any other Fed governor or president.
Notably, Logan includes a pointed critique of reserve quotas (or reserve tiering) that some critics of the ample reserves regime have floated a compromise. She frames them as antithetical to free-market principles, using the same libertarian lexicon that the advocates of a smaller balance sheet often deploy.
The tension to watch going forward is whether the demand-shifting approaches she outlines as a possible path forward can actually deliver enough balance sheet reduction to satisfy those who want a meaningfully smaller Fed footprint.
Trump’s protectionist Nixon redux takes a darker turn than the original
This is truly the outstanding column on the anniversary of Trump's April Fools' Liberation Day Tariffs. Must read by Alan Beattie in @financialtimes
https://t.co/tQg3yTTDU5
The danger for Turkey is always the same. The current government pushes growth way above the speed limit, which leads to a large current account deficit. Then a shock like high oil prices hits and things run out of control. We're at that point again...
https://t.co/7shtWiTjXb
The latest from @LibertyStEcon
Behind the ATM: Exploring the Structure of Bank Holding Companies
By Lily Gordon and Lee Seltzer
https://t.co/qiMKago6d9
On a recent Geopolitical Cousins episode I mentioned that BCA Research has opened its Gulf War dashboard to all users, including non-clients. Below is the promised link. Let us know if you'd want us to include some other data here.
https://t.co/x17XKa1fe4
Saudi Arabia, Russia, Iraq, UAE, Kuwait, Kazakhstan, Algeria, and Oman adjust production and reaffirm commitment to market stability
Read Press Release ➡️https://t.co/MEO3HaBieL