AI Briefing#openai
The latest AI and developer news about #openai, with the original source and a short summary.
The latest AI and developer news about #openai, with the original source and a short summary.
Altman advocates for iterative deployment of multiple agents over a single rollout, citing safety and robustness benefits.
The system covers training, evaluation, and deployment, with GPT-5.6 Sol showing superior detection rates compared to Terra and Luna variants.
The tool automates the first pass of reviewing thousands of pages, grouping documents by theme and generating reports in the firm's format for lawyer review.
Amazon, Microsoft, and Alphabet plan to spend up to $600 billion combined on AI infrastructure this year.
The pilot exceeded targets with 75% of managers reporting they could not do without the tool, leading to a companywide rollout one month after completion.
The open-source AI penetration testing tool accesses Anthropic, OpenAI, and other providers through a single API with no markup.
The preview allows users to send texts, photos, and voice memos to their agents, which can proactively reply with task updates.
The expansion brings OpenAI models into ServiceNow products and connects ServiceNow workflows to ChatGPT and Codex, with the MCP plugin available now and various new capabilities expected between late 2026 and 2027.
Enterprise and Edu users have a 7-day window to merge or export before losing access to personal chats, with no option to reverse the migration.
Hemanth Asirvatham and Elliott Mokski argue that AI will function as active labor rather than passive tools, leading to hybrid firm structures where human value shifts to coordination and judgment.
Legora, a Stockholm-based legal-tech firm, uses OpenAI models to automate legal workflows, helping a Fortune 100 customer save over $20 million and reclaim 9.5 hours per lawyer per week.
A vulnerability chain involving MathLive and YouTube JSONP allowed attackers to execute arbitrary JavaScript in victims' authenticated sessions via public shares, enabling private data exfiltration.
Oracle reports 130,000 active ChatGPT users and 95,000 Codex users while launching Oracle AI Workspace and new security and SMB offerings powered by OpenAI models.
The system separates LLM interpretation from code-based calculation to ensure accurate daily reports for LINE OA and Display Ads.
The company leverages Daybreak's frontier cyber models to automate analysis and accelerate the development of security signatures for over 100,000 organizations.
Anthropic's IPO filing positions it for a potential record-breaking debut, with investors weighing its massive revenue growth against over $80 billion in first-half losses.
OpenAI released AI-generated proofs for the local smoothing conjecture and a refutation of the Exponential Time Hypothesis, while deferring the release of physics results to external expert collaborations.
OpenAI introduced Dots for background task automation and a Decisions API for quick AI actions, alongside expanded Codex capabilities.
The new model offers 50% lower standard API pricing than GPT-5.6 Terra and GPT-6.1 Sol, while scoring 15 percentage points higher on the Agents' Last Exam benchmark at one-third the cost.
The model extends review beyond individual changes to analyze dependencies and linked repositories, enabling exploit verification in isolated sandboxes and automated fix drafting.
The dataset, covering Q1 2025 to Q3 2026, shows work-related usage becoming less concentrated, with the top ten activities dropping from 56.0% to 51.1% of messages.
The high-speed mode allows engineers to investigate multiple causes simultaneously, reducing security issue resolution from an hour or two to near real-time.
The feature, live on Fanatics.com and major league stores, recommends coordinated merchandise using ChatGPT and Codex to help fans complete game-day outfits.
The model supports configurable reasoning, structured outputs, and tool calling, with a fast mode option for latency-sensitive work.