Google DeepMind published a paper investigating the effects of training Gemma-4 4B with reinforcement learning to write moral arguments. The results showed that on ambiguous cases in MoralChoice, the proportion of answers that changed due to swapping the order of options dropped from 14.3% to 6.6%, but the proportion of sycophantic agreement with users only went from 45.3% to 44.0%, and accuracy on moral fable questions actually fell from 56.5% to 53.4%. Based on this, the paper recommends that when evaluating advice- or judgment-type tasks, consistency and correctness should be tested separately.
On October 30, 2026, Google DeepMind published the paper "The Economics of Intelligence," proposing that the economic impact of AGI be characterized by two thresholds—competence (single-task capability) and coordination (agents' ability to collaborate)—and arguing that the economic impact depends on agents' coordination capabilities rather than on individual models becoming stronger. The paper states that AI has already shown changes in lean startups, flattened organizational structures, and a slowdown in hiring for some entry-level positions, but that large-scale unemployment has not yet occurred; most benchmarks test only a single model, whereas firms compete as teams. The paper cites a 100-person test in which groups with complementary pairings performed 32% better than randomly paired groups. On the same day, Ethan Mollick commented on the research, pointing out that an organization is not merely the sum of its best individuals, that a single agent becoming stronger will not change the economy on its own, and that only agents capable of working in coordination may bring about transformation.
On October 30, 2026, Google DeepMind published the paper "The Economics of Intelligence," proposing that the economic impact of AGI be characterized by two thresholds—competence (single-task capability) and coordination (agents' ability to collaborate)—and arguing that the economic impact depends on agents' coordination capabilities rather than on individual models becoming stronger. The paper states that AI has already shown changes in lean startups, flattened organizational structures, and a s
On October 30, 2026, Google DeepMind published the paper "The Economics of Intelligence," proposing that the economic impact of AGI be characterized by two thresholds—competence (single-task capability) and coordination (agents' ability to collaborate)—and arguing that the economic impact depends on agents' coordination capabilities rather than on individual models becoming stronger. The paper states that AI has already shown changes in lean startups, flattened organizational structures, and a slowdown in hiring for some entry-level positions, but that large-scale unemployment has not yet occurred; most benchmarks test only a single model, whereas firms compete as teams. The paper cites a 100-person test in which groups with complementary pairings performed 32% better than randomly paired groups. On the same day, Ethan Mollick commented on the research, pointing out that an organization is not merely the sum of its best individuals, that a single agent becoming stronger will not change the economy on its own, and that only agents capable of working in coordination may bring about transformation.
Google DeepMind and several top universities have published a paper testing AI agents' ability to solve unpublished, research-level theoretical physics problems. According to the paper, the best-performing AI agent solved 49 of 50 unpublished research-level theoretical physics problems, each roughly equivalent to a day's work for a domain expert, at a cost of under $3 per problem. The problems were contributed by experts from their own research, and the agents answered independently, split into two groups—with and without internet access—and were graded solely on the final numerical answer. GPT-6-Astra, with internet access in Codex, solved 49; it and Opus 5.5, without internet access in Claude Code, each solved 46, with accuracy changing very little when internet access was removed, leading the authors to suggest that the models may have already memorized a large body of physics literature.
Google DeepMind and several top universities have published a paper testing AI agents' ability to solve unpublished, research-level theoretical physics problems. According to the paper, the best-performing AI agent solved 49 of 50 unpublished research-level theoretical physics problems, each roughly equivalent to a day's work for a domain expert, at a cost of under $3 per problem. The problems were contributed by experts from their own research, and the agents answered independently, split into
Google DeepMind and several top universities have published a paper testing AI agents' ability to solve unpublished, research-level theoretical physics problems. According to the paper, the best-performing AI agent solved 49 of 50 unpublished research-level theoretical physics problems, each roughly equivalent to a day's work for a domain expert, at a cost of under $3 per problem. The problems were contributed by experts from their own research, and the agents answered independently, split into two groups—with and without internet access—and were graded solely on the final numerical answer. GPT-6-Astra, with internet access in Codex, solved 49; it and Opus 5.5, without internet access in Claude Code, each solved 46, with accuracy changing very little when internet access was removed, leading the authors to suggest that the models may have already memorized a large body of physics literature.
Google DeepMind has published a paper studying the impact on employees of an AI assistant acting on its own without being asked. According to the paper, the agent ran for 5 months across more than 20 teams at a large tech company, and the researchers interviewed 17 users. It treated outdated documents as the latest content, flagged a VP on 16 bugs, shared a draft poster before the author was ready, and misread sarcastic tone. Some users called it a lifesaver, while others felt surveilled and moved sensitive discussions into chats without the agent; teams that actively chose the agent forgave its early mistakes, while those who felt it was imposed on them did not. The authors recommend obtaining team consent when deploying agents, separating permissions, and providing an easy way to undo actions.
Google DeepMind has published a paper studying the impact on employees of an AI assistant acting on its own without being asked. According to the paper, the agent ran for 5 months across more than 20 teams at a large tech company, and the researchers interviewed 17 users. It treated outdated documents as the latest content, flagged a VP on 16 bugs, shared a draft poster before the author was ready, and misread sarcastic tone. Some users called it a lifesaver, while others felt surveilled and mov
Google DeepMind has published a paper studying the impact on employees of an AI assistant acting on its own without being asked. According to the paper, the agent ran for 5 months across more than 20 teams at a large tech company, and the researchers interviewed 17 users. It treated outdated documents as the latest content, flagged a VP on 16 bugs, shared a draft poster before the author was ready, and misread sarcastic tone. Some users called it a lifesaver, while others felt surveilled and moved sensitive discussions into chats without the agent; teams that actively chose the agent forgave its early mistakes, while those who felt it was imposed on them did not. The authors recommend obtaining team consent when deploying agents, separating permissions, and providing an easy way to undo actions.
Google DeepMind, together with Google Research, the World Resources Institute (WRI), and the International Institute for Applied Systems Analysis (IIASA), released the Forest Typology (ForTy) 2020 global forest type map. The map classifies every 10×10 meter parcel of land worldwide into six forest and land-use types, and serves as a key baseline for the European Union's deforestation regulation. The team trained the model on about 1.7 million labeled locations, using AlphaEarth Foundations embedding vectors as a key input; in validation at more than 9,000 randomly sampled points, the forest type classification accuracy was 90%, and the accuracy in distinguishing forest from non-forest was 95%. A companion year-by-year time series from 2017 to 2025 is also openly available and can be used free of charge under an open license on Google Earth Engine. There are currently no further developments.
Google DeepMind, together with Google Research, the World Resources Institute (WRI), and the International Institute for Applied Systems Analysis (IIASA), released the Forest Typology (ForTy) 2020 global forest type map. The map classifies every 10×10 meter parcel of land worldwide into six forest and land-use types, and serves as a key baseline for the European Union's deforestation regulation. The team trained the model on about 1.7 million labeled locations, using AlphaEarth Foundations embed
Google DeepMind and Cambridge have published a paper demonstrating that a Gemini-based AI agent can autonomously discover new mathematics. The agent found and provided exact rules for 11 Busy Beaver Turing machines that had no prior analysis records. According to the paper, the agent first tested edge cases in a simulator, drafted informal proofs, and then completed the proofs in the Lean proof checker; all the relevant rules were verified through Lean. The paper states that this reflects AI moving from answering mathematical questions posed by humans toward open-ended discovery. Currently, this work is presented as a paper, and there have been no reports of further progress.
Google DeepMind and Cambridge have published a paper demonstrating that a Gemini-based AI agent can autonomously discover new mathematics. The agent found and provided exact rules for 11 Busy Beaver Turing machines that had no prior analysis records. According to the paper, the agent first tested edge cases in a simulator, drafted informal proofs, and then completed the proofs in the Lean proof checker; all the relevant rules were verified through Lean. The paper states that this reflects AI mov
The Google DeepMind team had four AI agents running on Gemini 4 Argon autonomously complete Minecraft in a randomly generated world, killing the Ender Dragon with no human input throughout. The team did not assign them roles, strategies, or game-specific tools; the four agents divided the work among themselves, warned each other of threats, recovered items from fallen teammates, and recorded mistakes in a shared memory. The entire run took 27 hours and 14 minutes, during which they wrote a total of 2,783 JavaScript programs, exchanged 229 chat messages, and died 605 times, costing $1,620 at the standard Gemini 4 Argon pricing that takes effect in 2027.
The Google DeepMind team had four AI agents running on Gemini 4 Argon autonomously complete Minecraft in a randomly generated world, killing the Ender Dragon with no human input throughout. The team did not assign them roles, strategies, or game-specific tools; the four agents divided the work among themselves, warned each other of threats, recovered items from fallen teammates, and recorded mistakes in a shared memory. The entire run took 27 hours and 14 minutes, during which they wrote a total
On October 23, 2026, Google DeepMind released the frontier model Gemini 4 Argon. Google's official blog stated that the model performs outstandingly on complex tasks such as coding, mathematics, and 3D world building, and surpasses other frontier models on benchmarks across multiple domains; the company also showcased five community creative projects, including a 3D Earth that maps wildfires and seismic waves in real time, a turbofan engine simulation with live data, a 3D horse-drawn carriage generated from a single image, and a multiplayer capture-the-flag browser game, and said the model had already been made broadly available last week. On the same day, another report stated that Argon can learn physical tasks by watching videos and transfer them to Apptronik's Apollo humanoid robot, with Apollo hitting seven consecutive shots in internal tests; the report also said Argon scored 87% on the embodied reasoning benchmark ERQA. On October 24, the Gemini for Android app pushed a redesign of its sidebar and settings interface via a server-side update, and reports said the redesign had already arrived on iOS, macOS, and Web alongside Gemini 4 Argon last week.
On October 23, 2026, Google DeepMind released the frontier model Gemini 4 Argon. Google's official blog stated that the model performs outstandingly on complex tasks such as coding, mathematics, and 3D world building, and surpasses other frontier models on benchmarks across multiple domains; the company also showcased five community creative projects, including a 3D Earth that maps wildfires and seismic waves in real time, a turbofan engine simulation with live data, a 3D horse-drawn carriage ge
On October 23, 2026, Google DeepMind released the frontier model Gemini 4 Argon. Google's official blog stated that the model performs outstandingly on complex tasks such as coding, mathematics, and 3D world building, and surpasses other frontier models on benchmarks across multiple domains; the company also showcased five community creative projects, including a 3D Earth that maps wildfires and seismic waves in real time, a turbofan engine simulation with live data, a 3D horse-drawn carriage generated from a single image, and a multiplayer capture-the-flag browser game, and said the model had already been made broadly available last week. On the same day, another report stated that Argon can learn physical tasks by watching videos and transfer them to Apptronik's Apollo humanoid robot, with Apollo hitting seven consecutive shots in internal tests; the report also said Argon scored 87% on the embodied reasoning benchmark ERQA. On October 24, the Gemini for Android app pushed a redesign of its sidebar and settings interface via a server-side update, and reports said the redesign had already arrived on iOS, macOS, and Web alongside Gemini 4 Argon last week.
Google DeepMind, together with Google Research, the World Resources Institute (WRI), and the International Institute for Applied Systems Analysis (IIASA), released the Forest Typology (ForTy) 2020 global forest type map. The map classifies every 10×10 meter parcel of land worldwide into six forest and land-use types, and serves as a key baseline for the European Union's deforestation regulation. The team trained the model on about 1.7 million labeled locations, using AlphaEarth Foundations embedding vectors as a key input; in validation at more than 9,000 randomly sampled points, the forest type classification accuracy was 90%, and the accuracy in distinguishing forest from non-forest was 95%. A companion year-by-year time series from 2017 to 2025 is also openly available and can be used free of charge under an open license on Google Earth Engine. There are currently no further developments.
Google DeepMind and Cambridge have published a paper demonstrating that a Gemini-based AI agent can autonomously discover new mathematics. The agent found and provided exact rules for 11 Busy Beaver Turing machines that had no prior analysis records. According to the paper, the agent first tested edge cases in a simulator, drafted informal proofs, and then completed the proofs in the Lean proof checker; all the relevant rules were verified through Lean. The paper states that this reflects AI moving from answering mathematical questions posed by humans toward open-ended discovery. Currently, this work is presented as a paper, and there have been no reports of further progress.