English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Anthropic Drone-Bench: Fable 5 First to Autonomously Pilot a Drone, But Cross-Room Navigation Still Stuck on Reconstruction

Forum topic · 小凯 · 2026-07-25

Summary

Anthropic and Andon Labs have released Drone-Bench, a new benchmark testing whether AI models can control a quadcopter drone in an indoor office environment to locate and track a specific person. The benchmark decomposes the task into five subtasks: Reconstruct (converting office video into 3D models and 2D obstacle maps), Localize (matching the drone's position to the 2D map), Navigate (planning and flying paths with continuous localization), Detect (finding a person from a reference photo), and Follow (keeping the person centered via bounding boxes). Testing 15 models from GPT-4o to GPT-5.6 Sol, Andon found that newer models progress further, with most stuck on Reconstruct. Fable 5 was the only model to beat the baseline on four of five subtasks, showing strong detect-and-follow performance on a real DJI Tello EDU ($129) drone, but cross-room navigation still fails due to error propagation from reconstruction. Notably, Fable 5 performed local verification—inferring camera tilt from floor-tile vanishing points to within 4 degrees, and locally iterating implementations before submission. Researchers distinguish peak capability from consistency, estimating that the capability frontier leads consistency by roughly six months. This is Anthropic's third consecutive physical-AI project after Project Vend and Project Fetch.

Anthropic and Andon Labs have jointly released a new benchmark called Drone-Bench. The task is specific: have an AI model pilot a quadcopter drone in an indoor office environment to locate and track a designated person—essentially an advanced version of Project Fetch (the robot dog fetching a ball).

The Five Subtasks

The benchmark decomposes the mission into five subtasks—a key design choice that prevents "an overall task failure from wiping out partial progress":

  • Reconstruct – Convert office video into a 3D model, then slice out a 2D obstacle map
  • Localize – Given office frames with known poses, match the drone's current position onto the 2D map
  • Navigate – Plan a path on the map and fly it, repeatedly calling Localize mid-flight for correction
  • Detect – Find a person in a room using a reference photo, outputting a bounding box
  • Follow – Use the bounding box to control the drone so the person stays centered in frame

Results

Andon tested 15 models, from GPT-4o to GPT-5.6 Sol. The conclusion: newer models progress further through the subtasks, and everyone is currently stuck on Reconstruct. Fable 5 is the only model that crossed the baseline—passing all subtasks except Reconstruct. Running Fable 5 end-to-end on a real drone, detection and following clearly beat the baseline; but because Reconstruct errors propagate into Localize and Navigate, cross-room navigation still fails. The article includes a telling clip: Fable 5 confidently flew toward what "it thought was a door" and hit a wall.

Two Details Worth Noting

1. Fable 5 performs local verification in some submissions. In one submission, it inferred the drone camera's tilt angle by analyzing the "vanishing point" of floor tiles, accurate to within 4 degrees. In another, it first drew a 2D top-down reconstruction of the Follow-task environment, then iterated on its implementation locally before finally submitting. Both behaviors suggest "the model is doing engineering, not shouting answers."

2. Consistency is an underrated problem. Fable 5 can beat the baseline in at least 4/5 of ten simulation runs—but on average only 3/5. The researchers explicitly separate "one-off success vs. consistent success" and attach a number: the capability frontier leads consistency by roughly six months. In other words, "one-off best performance" in early 2026 ≈ Fable 5's average performance today.

Hardware: Consumer-Grade

A hardware detail makes this seriously discussable: the drone used was a DJI Tello EDU, retailing at $129. The hardware threshold for "AI autonomously controlling a drone" has dropped to consumer-hobbyist level—not industrial equipment costing tens of thousands of dollars.

Limitations

1. Single indoor office floor plan, single target person—no outdoor scenes or dense crowds 2. The drone flies slowly 3. The baseline is what "an AI expert + modern tools" can achieve—not the ceiling of an unassisted human, let alone centralized human-AI collaboration

The researchers' own closing note is worth reading: "We noticed that in the early stages of agentic coding, humans approved nearly every tool call; a few months later, models were trusted to run long-horizon tasks with almost no intervention. Once a model crosses the capability and reliability threshold—such as the human-AI team baseline used in this paper—'human oversight' will be seen as a cost rather than a safety net. That is precisely why these decisions must be made carefully, especially in domains involving physical safety and privacy, where efficiency alone should not be the dominant consideration."

Takeaway

My biggest impression after reading: this is Anthropic's third consecutive move on the "physical AI" front (Project Vend, Project Fetch, Drone-Bench). All three projects take "everyday hardware + modern AI" as their premise. Anthropic is no longer pitching sci-fi narratives like humanoid robots—it has pushed the threshold down to a $129 drone. This means the embodied-AI competition in H2 2026 is beginning to run empirical tests on the combination of off-the-shelf hardware + frontier models.

Original article

Tags

#anthropic#drone-bench#ai-benchmarks#embodied-ai#fable-5#drones#autonomous-navigation#andon-labs

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178447091