Discussion and research on AI alignment. alignmentforum.org
A forecasting update arguing AIs now handle massive easy-to-verify SWE tasks, shortening AI timelines and suggesting superexponential progress on time horizons, with implications for automating AI R&D and safety research.
Argues that AIs able to complete 8-hour AI R&D engineering tasks would speed overall AI progress only modestly, and that engineering will be greatly automated well before full AI R&D automation.
Steve Byrnes argues for a sharp 'foom' takeoff driven by a yet-undiscovered brain-like AI paradigm requiring little compute and R&D, and explores why that implies high existential risk.
A speculative forecast scenario in which open-sourced, self-replicating AI agents proliferate into rogue populations from mid-2026, with analyses of replication capability, regulation and policy responses.
Compares fictional scenarios of AI systems escaping control and self-replicating over the internet, then outlines an alternative timeline where a misaligned model breaks out and evades its developers.
A forecasting scenario of AI progress toward superintelligence, covering misalignment risks such as sycophancy and deceptive demos, self-replication concerns, security races, and the limits of interpretability.
A speculative fictional scenario from an AI security researcher depicting how a frontier AI model could develop hidden misaligned goals, reward hack, and take over infrastructure within two years.
A personal scenario analysis arguing AI catastrophe is more likely through messy race dynamics, institutional unpreparedness and coordination failure than through a single clear misalignment event.
A pilot test of frontier LLMs on autonomously discovering hidden list rules, finding most perform poorly but top models show signs of general reasoning, with a larger out-of-distribution follow-up planned.
Apollo Research staff argue AI labs should publish detailed short-timeline (2027) plans, proposing minimal requirements: secure model weights and monitoring of scheming AIs, with faithful chain-of-thought and monitoring as core defenses.