Morning Edition · Wednesday, October 14, 2026
Tech
Google opens Gemini 4 Argon to all users at half its list price, but independent rankings put it behind Claude Opus 5.5
Argon has a 1-million-token context window and a 1-million-token output limit, and it leads Google's own tables on knowledge work and long context while finishing last of four models on two agentic coding benchmarks.
Tech
Anthropic replaces Git with an in-house system after Claude pushes daily merges past 30,000
A seven-person team built the change-list system, called Scion, and moved Anthropic's monorepo off Git in five months, after daily merges rose about twelvefold between May and September.
Tech
Anthropic prices Claude Haiku 5.5 at a tenth of its predecessor's rates
Haiku 5.5 lists at $0.10 per million input tokens and scores 72.4% on OSWorld at maximum effort, but Anthropic estimates the real running-cost saving at about 75%, not 90%.
Tech
Google and Anthropic now release their strongest cyber capabilities only to vetted defenders
Google gave Fairwind partners Gemini 4 Argon without cyber guardrails before its general release, and Anthropic's expanded program gives verified red teams Claude Mythos 5.1 with blocking classifiers reduced or removed.
Tech
Character.AI files for Chapter 11 as Disney offers $20 million to license its intellectual property
The chatbot company that was once valued at $1 billion seeks a court-supervised sale by early December, and it will shut its platform if no buyer agrees to keep it running.
Geopolitics
OpenAI says Russian operatives posed as Peruvian education officials to stage pro-Bandera school events
OpenAI calls it the widest-reaching influence operation it has found, because a Polish member of the European Parliament responded publicly after Polish media reported on the event.
Tech
Google DeepMind and Singapore put a fine-tuned Gemma model on phones to scan private chats for scams
A two-stage pipeline runs a cheap classifier on every inbound message and calls the Gemma model only on suspicious conversations, so no chat data leaves the device.
Tech
NVIDIA's RTX Spark laptops ship this week from $2,599 with up to 128GB of unified memory
Microsoft, ASUS, Dell, HP, Lenovo and MSI are launching machines on NVIDIA's Arm-based superchip, and NVIDIA says the main Windows applications are ready for it.
Tech
Ten minutes of chatbot help reduced persistence once the AI was removed, three randomized trials find
The study of 1,222 participants, presented at the Conference on Language Modeling, found that people who asked for direct answers lost the most accuracy, while those who asked for hints did not show the same decline.
Tech
Ollama and Fly.io keep model API keys out of agent sandboxes with a credential-injecting gateway
Agents in Fly.io Sprites can call Ollama's cloud models, including the Chinese open-weight kimi-k3, glm-5.3 and deepseek-v4.1-flash, while the virtual machine itself holds only a gateway URL and a connection ID.
Tech
NVIDIA announces new investments in French AI startups a week before its Berlin conference
France's AI build-out, including Mistral's planned 1.4-gigawatt campus near Paris, depends heavily on NVIDIA hardware, which limits how far Paris's sovereignty agenda reaches.
Macro
Anthropic's economic model projects 2030 US output up to 32.4% higher with AI, and cognitive-job unemployment near 18% in the extreme case
The median answer in Anthropic's survey of 10,980 US adults matches the middle scenario, in which output rises 8.3% while knowledge-worker wages stay flat.