Six AI debates: 1. Frontier models get cheaper vs. more expensive 2. The world is underinvested vs. overinvested in compute 3. The gap between frontier and open source shrinks vs. widens 4. AI is job-destroying vs. job-creating 5. AI detection & regulation keeps pace with generation vs. never catches up 6. AGI/ASI is in sight vs. unachievable Answers vary heavily based on age, job, industry, time-spent-on-Twitter.

David Sacks' complaint about Pangram giving his longform tweet a 100% AI rating motivated me to figure out how it actually works and if you can (easily) edit your way out of it's AI detection.

LinkedIn users have clicked the platform’s new “Seems like AI slop” button more than 1 million times. LinkedIn added the option so people can flag posts they think are low-quality AI-generated content. LinkedIn says the feedback, along with its AI detection systems, has helped reduce views of content it identifies as AI-generated spam by 40% in recent weeks. One analysis found that 41% of the LinkedIn long-form posts it checked were flagged as fully AI-generated.

Day 1 of #FalCon2026# started with a look at what’s next for AI, cybersecurity, and the agentic era. CrowdStrike CEO and founder @George_Kurtz took the stage alongside @NVIDIA founder and CEO @JensenHuang , @intel CEO @LipBuTan1, and @OpenAI co-founder and President @gdb to explore how AI is reshaping technology and why security will be foundational to what comes next. And George unveiled major new innovations built to secure the AI era: 🛡️ Falcon Guardian: AI Detection and Response that secures AI agents at runtime, where they execute. Guardian brings discovery, visibility, access controls, detection, and response across the AI estate. 🧠 CrowdStrike SafeMind: Purpose-built offensive and defensive AI models, created with NVIDIA, that work together to find weaknesses, strengthen defenses, and help defenders operate at machine speed. 🔬 Cyber Superintelligence Lab: A new frontier AI research organization bringing together CrowdStrike AI researchers, offensive operators, and incident responders to advance cyberdefense and AI safety. The AI era is accelerating. So is the mission to secure it. And we’re just getting started at #FalCon2026#.

The Gavin Baker vs. Sholto Douglas exchange on the economic concentration of power of AI is one of the best AI debates. Here's a gist with some context and some thoughts: One of the central themes of the debate offense-defense balance and how it governs whether AI is (or should be) concentrated in the hands of few or distributed to many. The core idea behind offense-defense balance, outlined in Robert Jervis's 1978 essay "Cooperation Under the Security Dilemma" on world politics, is: for a given technology, is it cheaper to attack or to defend? Baker initially argues that he agrees with Zuck's view in his essay that AI should be distributed and not concentrated into a few corporations' hands. Sholto rebuts that it squarely depends on the offense-defense balance of the domain (cyber, bio, etc). And even if you do distribute, the bottleneck is compute. He cites two domains: — Cyber he says is defense-dominant, eventually. If it's opened up today, the cost to attack is minimal and the you cannot defend quickly enough. We'll need a ~2yrs of hardening the critical systems like financial institutions and infra providers before this may no longer be a risk. — Bio he says is offense-dominant well into the 2030s. Synthesizing deadly self-replicating pathogens could cost O($10K) and vaccines cost billions and years. You just can't patch the human body with smarter models. The distribute everything vs gate everything is a false dichotomy. You want to distribute in defense dominant domains and gate in offense dominant. And its all extra muddied because capabilities are dependent on compute which is extremely unequally distributed. And compute eventually dictates proliferation of capabilities. ______________ Both gloss over some offense dominant domains that are already distributed massively: — Propaganda and persuasion, for example, were once regarded offense dominant yet already ubiquitous today. That said, technology changes, and watermarking outputs and AI detection tools may have shifted it to being defense dominant. — Mass surveillance is also offense dominant (usually the attacker is the state) and most LLMs you can use for <$1/M toks can do this today. Today, even non-state actors (private cos) can surveil you in pretty dangerous detail by running agent swarms on vast amounts of public data. Both Gavin and Sholto are right, and we see it today. — Sholto's right: the very pinnacle of frontier capabilities is locked away in those who have compute and will continue to be. The shape of compute of the >100 GW compute by 2030 is almost known at this point. Between 6-7 US entities (2 big labs, 3 hyperscalers clouds, Meta/xAI) will own or control 65-75%. With RSI around the corner, as long as it holds true that more compute leads to being able to train and serve more powerful models at scale, those companies will have proprietary frontier capabilities and only they / the market will dictate when they need to release them. Notably, this isn't monopolistic concentration of power since multiple parties will have that capability. — Gavin's right too: frontier capabilities in certain domains are making it to the public and will continue to do so. Models are getting cheaper. Whether we like it or not, people can do dangerous things with models today. LLMs already cyber exploits that are less talked about (take even the LLM-assisted hack on the nationwide Indian exam system, CBSE). As long as prices for models continue to fall at a more rapid rate (10x YoY) than growth in compute (2-3x YoY), this dynamic will continue. Frontier intelligence will continue to be more accessible.

Live from our biggest #FalCon# ever! 🚀 CrowdStrike CEO George Kurtz joined Jim Cramer on @CNBC's Mad Money to discuss how CrowdStrike is securing the AI era. From introducing Falcon Guardian and defining AI Detection and Response (AIDR), to unveiling SafeMind with NVIDIA technology, CrowdStrike is delivering the innovation enterprises need to secure AI at scale. “AI is a transformational technology. We have to use it. It just has to be secure.” — George Kurtz 🔗 Tune in:

On 31st August the US Defense Department denied that America had struck Iran's main oil export terminal. It was correcting the President of the United States, who had posted video of that strike the night before. President Trump published an AI generated clip captioned Kharg Island being blown to smithereens, posted it again without the caption, then added a second clip of missiles hitting a coastal oilfield. Reuters ran AI detection software over the footage and found it most likely synthetic. Kharg handled 90 percent of Iran's crude exports before the war and houses about 20,000 people, most of them oil workers. Hamid Bovard, who runs the National Iranian Oil Company, called the post laughable and said conditions were calm, with crews busy repairing earlier damage. The real operation happened 660 kilometres east. CENTCOM struck two rocket launchers on Larak Island, saying Revolutionary Guard minelaying forces posed an imminent threat to the strait. Iran answered before dawn with ballistic missiles at the King Hussein and Al Azraq air bases in Jordan. Jordan's armed forces say they destroyed eight before anything threatened people or property. The UAE denied Iran's claim on Al Minhad while confirming it engaged a drone over its waters. Iran's claims outran its results too, and the difference matters. The missiles and the drone were real. What is disputed is what they hit. The video showed destruction at a site the Pentagon says was never attacked. Brent rose about two percent in thin holiday trading, which tracks the launchers and the missiles rather than the footage. Every statement from this war now has to be split three ways. What was launched. What was hit. What was only shown. He can still order the strike. The post cannot confirm it.