Anthropic And OpenAI: "Pacing The Frontier" Or Running Out Of Steps?

Ignacio Planas Gonzalez 305 Followers Follow Summary Anthropic and OpenAI's calls to 'pace the frontier' hinge on claims of imminent recursive self-improvement (RSI), yet current evidence suggests RSI is not here yet. Anthropic's own productivity evidence shows agents optimizing human-defined problems faster, not choosing the problems. Every capability step so far, from transformers to reasoning to agents, was a human discovery. Frontier labs' economic moat is threatened by model commoditization and distillation, especially if regulatory pacing slows innovation while open models rapidly close the gap. Without RSI, AI utility will grow but value capture may shift away from frontier labs, making future margins and competitive advantages uncertain once these companies IPO. Getty Images Anthropic Says It's Time To Pace The Frontier, OpenAI Agrees Following a flurry of news regarding AI safety, including some high-profile resignations, cybersecurity incidents involving "agent swarms," and much speculation around regulatory capture, Anthropic ( ANTHRO ) CEO This article was written by Ignacio Planas Gonzalez 305 Followers Follow As an entrepreneur, software engineer and economist, I have always had a passion for investing. The feeling of being ahead of everyone else and being rewarded with outsized, risk-adjusted returns is something that has drawn me to the stock market since I was old enough to start investing. I hold an economics degree from the University of Barcelona and have attended professional-level masterclasses on portfolio management, options trading and macro analysis taught by ex-Goldman Sachs, JP Morgan and Bank of America traders. My strategy involves long/short options portfolio management, with a 1-3 month time horizon. The main things I look for in my longs are growth, operational leverage and growing TAM as the main criteria. The main thing I look for in my shorts, is the exact opposite: deteriorating fundamentals, margin compression, shrinking TAM and valuation disconnect. I also manage a long term, long-only portfolio. There, dividend growth, cash generation, sensible valuation and MOAT defensibility are the main focus. The fields where I feel I have an edge on are technology, SaaS, risk management, and options strategies that squeeze every last dollar out of a good idea. In the age of AI, many opportunities are opening up in unexpected places, and I feel like my background as a software engineer and entrepreneur who has seen both sides of the AI coin gives me a unique perspective. I have watched AI create amazing efficiencies for the companies I have created and the ones that I observe in the market, but I have also seen it destroy MOATs and compress valuations immensely. I think these articles will be highly beneficial to the readers, because a lot of work will be put on them regardless of the outcomes. Writing my ideas down and spelling them out loud has always increased my conviction and made me see them through to the end, so this will also be helpful for myself. I will make sure to generate the highest possible quality articles, and my pride and reputation being on the line will make me more strict with myself and the quality of my ideas. Analyst’s Disclosure: I/we have no stock, option or similar derivative position in any of the companies mentioned, and no plans to initiate any such positions within the next 72 hours. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article. Seeking Alpha's Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.