Inflection Points, AI’s Sydney

Inflection Points

There is no shortage of bear market theories circulating in the investment community and media. While there are many intelligent reasons to be negative, there comes the point where reflexive bearishness turns into a blindspot. It seems likely to me that we are at that point today. There are a handful of key inflection points from recent months that demonstrate how key underlying trends have changed course.

• Currencies: the DXY US Dollar index peaked in September when the Euro, GBP, and Yen hit panic-level prices. It would be a major surprise to revisit $.95 on the Euro (currently $1.07) and $1.06 on the GBP (currently $1.20). It’s a bumpy road up and a bumpy road down, but the trend remains. That strong dollar was starting to put immense pressure on international economies that consume US dollar goods.
• Commodities: Many critical commodities are well past their extreme panic-level prices. Lumber and international shipping are not returning to their previous highs without another Covid style disruption. On the other hand, steel is still stuck at its high and labor inflation remains elevated. The evidence shows that parts of inflation were transitory. Even Lithium is beginning to fall a bit.
• Rates: The 10 year treasury peaked in October at 4.24%. Short-term rates remain high while the Fed continues to raise their interest rate. The next hike on March 22nd will likely be .25%. The following hike is 2.5 months away, which is plenty of time for lagging inflation data to roll in. By May, many YoY price changes will be negative for the first time in years.
• Time: Even though it feels like yesterday, it’s been two years since the peak of the GameStop mania and 15 months since the Nasdaq peak. In the last major crisis in 2008 the Nasdaq fell for 18 months.

Baupost, famous as the margin-of-safety value investor, recently added Meta, Google and Amazon shares to their portfolio. While 13F information is not actionable, outdated and paints an imperfect picture at best, those three companies are among the top holdings of the Nasdaq index and not typically present in shrewd value investor portfolios.

To sum up, there can always be more volatility, but the bulk of the money made by this crop of short sellers is likely well behind us.

AI Sydney’s Emergent Behavior

It’s difficult not to write about AI each week, and I promise to keep it relevant. Ben Thompson’s piece yesterday on Bing’s “Sydney” stood out as worthy of a highlight. Ben is famous for his long-running and well-respected technology blog, Stratechery.

In the article, Ben describes his interactions with Bing’s AI program, “No, I don’t think that Sydney is sentient, but for reasons that are hard to explain, I feel like I have crossed the Rubicon. My interaction today with Sydney was completely unlike any other interaction I have had with a computer, and this is with a primitive version of what might be possible going forward.”

The presence of personality in LLMs (large language models) is an emergent phenomenon. Given that LLMs are trained on content from the internet, it can essentially pull from the nuances of personalities in the underlying data and via the people tasked with tweaking the model. In this case the personality of “someone who is under-appreciated and over-achieving and constantly feels disrespected” is present in Ben’s interaction.

Bing AI’s ability to demonstrate personality and communicate emotion is less about disrupting Google’s search business and more about the evolution of an entirely new technology. If you haven’t seen the movie Her, I highly recommend it to get a sense of how this can evolve from here. The way we communicate with computers to gather information and even for entertainment is changing radically.