OpenAI is building a unified desktop application that combines ChatGPT, its Codex AI coding tool, and the Atlas AI-powered browser. The consolidation aims to reduce product fragmentation that OpenAI leadership says has slowed development and quality standards.
Google is testing a native Gemini app for macOS, according to Bloomberg. The app would compete directly with OpenAI's ChatGPT and Anthropic's Claude, both of which offer standalone Mac applications. A key differentiator is 'Desktop Intelligence,' which allows Gemini to view screen context and pull content from apps to personalize responses.
ElevenLabs has launched a music marketplace where users can publish and sell tracks created with its ElevenCreative AI music model, with the company claiming to have already generated nearly 14 million songs on the platform. The company has paid out over $11 million through its Voice Marketplace using the same model. However, AI-generated music lacks legal copyright protection, leaving all legal risk on users.
Google has begun beta testing a dedicated Gemini app for macOS with select users, according to Bloomberg. The early version includes only critical features and hints at a "Desktop Intelligence" capability that lets Gemini see screen context. The move addresses a competitive gap, as Anthropic and OpenAI already offer native Mac apps for Claude and ChatGPT.
Cursor released Composer 2, a code-specialized model priced at $0.50 per million input tokens and $2.50 per million output tokens—roughly 90% cheaper than Claude Opus 4.6 ($5.00/$25.00) and 60% cheaper than GPT-5.4 ($2.50/$15.00). The model scores 61.3 on Cursor's internal CursorBench, competitive with Claude Opus 4.6 (58.2) but below GPT-5.4 Thinking (63.9).
Amazon's Alexa+, its conversational AI assistant, begins Early Access in the UK on March 19, rolling out to hundreds of thousands of users. The updated assistant includes British English localization with regional speech patterns, slang understanding, and agentic capabilities. Free for Prime members; £20/month for non-members.
Adobe is rolling out Firefly Custom Models in public beta, allowing creators to train AI image generators on their own artistic assets. The tool preserves visual consistency across character designs, illustrations, and photography styles without exposing training data to Adobe's general models.
Amazon is rolling out Alexa+, its AI-powered conversational assistant, to the UK as the first market outside North America. The company is offering free early access to new Echo buyers and plans to reach hundreds of thousands of customers in the coming weeks. After the beta period ends, Prime subscribers will get free access while non-Prime users will pay £19.99 per month.
Okta has released Okta for AI Agents, a management platform that addresses three core requirements: discovering deployed agents, monitoring their activities and access permissions, and terminating agent access when needed. The platform integrates with Salesforce, ServiceNow, Google, and AWS to import agents and their metadata, while providing continuous background scanning for unmanaged agents.
Google is testing a UI redesign for its Gemini app that moves the Temporary chat feature from a navigation drawer icon to a prominent button on the homepage. The change aims to make privacy-focused conversations more discoverable while maintaining Temporary chats' core privacy guarantee: conversations aren't stored in history, used for model training, or personalization.
Google Deepmind has expanded the Gemini API to enable multi-tool chaining, allowing developers to combine built-in tools like Google Search and Google Maps with custom functions in a single request. Results from one tool now automatically pass to the next through context circulation, eliminating the need for separate sequential handling.
Perplexity has released Comet, its AI-powered browser, as a free standalone app for iPhone users. Originally launched on PC at $200 per month, the iOS version joins recently-released Android and existing Windows and Mac versions. The browser combines web browsing with AI assistance for summarization, research, and task automation.
Google has expanded Personal Intelligence, its hyper-personalized Gemini mode, from $20/month subscribers to all users. The feature integrates data from Gmail, Search history, Google Photos, and other Google services to provide contextual assistance, though it remains entirely opt-in.
OpenAI released GPT-4o mini on July 18, 2024, a compact multimodal model with 128,000 token context window priced at $0.15 per million input tokens and $0.60 per million output tokens. The model achieves 82% on MMLU and claims to rank higher than GPT-4 on chat preference leaderboards while costing 60% less than GPT-3.5 Turbo.
Amazon Nova 2 Lite expands context window to 1 million tokens, introduces extended thinking with developer controls, and adds native tool use and web grounding. AWS claims Nova 2 Lite surpasses Nova 1 Pro on multi-step reasoning while costing 7x less and running up to 5x faster.
Midjourney has released an early version of V8 for community testing, achieving roughly 5x faster image generation and introducing native 2K resolution via --hd mode. However, premium features including --hd, --q 4, style references, and mood boards cost four times as much as standard generation, with Relax mode unavailable at launch.
Meta's recently acquired AI startup Manus launched a desktop application enabling its AI agent to directly access local files, tools, and applications on personal computers through a 'My Computer' feature. Previously cloud-only, the move positions Manus to compete with OpenClaw, the open-source AI agent that sparked recent industry momentum. Unlike OpenClaw's free, MIT-licensed offering, Manus operates as a paid subscription service.
DuckDuckGo's Duck.ai chatbot platform now includes OpenAI's GPT-5 mini for free users and GPT-5.2 for subscribers, both with reasoning capabilities. The platform continues to anonymize all conversations by default, stripping metadata before routing chats to model providers including Anthropic, Meta, Mistral, and OpenAI.
OpenAI has released GPT-5.4 mini and GPT-5.4 nano, compact models optimized for coding and subagent tasks. The new models deliver significant performance improvements—GPT-5.4 mini reaches 54.4% on SWE-Bench Pro versus 45.7% for GPT-5 mini—but cost 3-4x more per input token than their predecessors.
Mistral has released Leanstral, a 120B-parameter code verification agent built with the Lean programming language, claiming it outperforms larger open-source models and offers significant cost advantages over Anthropic's Claude suite. The model achieves a pass@2 score of 26.3—beating Claude Sonnet by 2.6 points—while costing $36 to run compared to Sonnet's $549.