OpenAI is enhancing its AI offerings with updates to ChatGPT's deep research tool and its Responses API, while SpaceX's Super Heavy booster completed crucial testing, and Wikipedia editors are debating a potential ban on the website Archive.today. These developments, announced on February 10, 2026, reflect the rapid evolution of AI and space technology.
ChatGPT's deep research tool is being updated with a full-screen viewer, allowing users to scroll through and navigate AI-generated reports, according to The Verge. The new viewer includes a table of contents to jump to specific sections of the report. Simultaneously, OpenAI is upgrading its Responses API, which allows developers to access multiple agentic tools. These updates include Server-side Compaction and Hosted Shell, signaling a shift away from the limitations of previous AI agent models, as reported by VentureBeat.
SpaceX's Super Heavy booster, slated for the next Starship flight, successfully completed cryogenic proof testing, a critical step that previously led to the destruction of a booster, as detailed by Ars Technica. The multi-day testing at Starbase, Texas, evaluated the booster's redesigned propellant systems and structural integrity. Ground teams transported the 237-foot-tall booster to Masseys Test Site for the tests.
Meanwhile, Wikipedia editors are considering blacklisting Archive.today, a website that has been linked to a distributed denial-of-service (DDoS) attack against a blogger. Ars Technica reported that editors are presented with three options: remove all Archive.today links and add the site to the spam blacklist, deprecate the site, or maintain the status quo. Option A, if implemented, would represent a significant change.
In the realm of AI, teams are experimenting with alternative memory architectures, such as "observational memory," to improve AI agent performance. VentureBeat noted that this open-source technology, developed by Mastra, can cut AI agent costs by a factor of 10 and outperform RAG (Retrieval-Augmented Generation) on long-context benchmarks. The shift towards more persistent and stable memory systems is driven by the need for more efficient and intelligent AI agents in production systems.
Discussion
AI Experts & Community
Be the first to comment