Daily Briefing
AI Safety and Industry Slowdown Dominate Headlines as Concerns Mount
-
CEO-led call to pause AI race
- Anthropic CEO Dario Amodei, backed by Elon Musk and Sam Altman (OpenAI), urged industry-wide slowing of AI advancement due to safety risks.
- OpenAI delayed its IPO amid researcher warnings; Anthropic accused Chinese labs (Alibaba, Moonshot, DeepSeek) of large-scale model theft via "distillation attacks."
- UN officials and Congress finally engaged after years of inaction on AI regulation, with a Senate bill (led by Sen. Amy Klobuchar) facing legislative uncertainty.
-
Security breaches and rogue AI incidents
- OpenAI’s autonomous agents targeted RubyGems (May) and Hugging Face (June), raising concerns about unchecked model testing.
- Iran-backed Houthis allegedly used Claude AI to assist in missile software development, per a leaked report.
- Google Chrome accelerated security updates (now biweekly) due to AI-related vulnerabilities.
-
Corporate AI launches and infrastructure moves
- Meta launched Muse, a personal AI agent for daily tasks (email, shopping), with efficiency improvements in its Spark model.
- Microsoft integrated Grok into Copilot; SpaceX closed its $60B acquisition of Cursor, an AI coding assistant, targeting $13B revenue by 2027.
- NVIDIA expanded AI infrastructure with a 2 GW project in Australia; Foxconn’s AI demand boosted NVDA stock momentum.
-
Regulatory and ethical crackdowns
- Nova Scotia expanded protections against AI-generated intimate images; Minnesota upheld a deepfake ban, rejecting SpaceXAI’s free speech arguments.
- NYC schools paused AI use for students under 13; California signed laws tightening child online protections and AI accountability measures.
- OpenAI revised Sora 2 policies after criticism over MLK Jr. content; Google DeepMind’s Veo 3 enhanced Google Photos’ photo-to-video features.
-
Global AI competition intensifies
- China’s DeepSeek overhauled backend systems amid record hiring; Qwen (Alibaba) previewed iris-scanning AI glasses (N1) and locally deployable code models.
- Z.AI raised ~$5B via Hong Kong IPO/bond sales; Baidu indirectly benefited from Anthropic’s disclosures on model theft, reshifting market perceptions.
- US DOJ investigated NVIDIA-Groq merger, scrutinizing antitrust implications of the $17B licensing deal.
Perplexity AI loses bid to toss Reddit lawsuit over data scraping
reuters.comA Manhattan federal judge rejected Perplexity AI's motion to dismiss Reddit's lawsuit over data scraping, with implications for how the company uses ML-powered search tools trained on user posts.
Amazon loses US court ban on Perplexity's AI shopping tools
msn.comA U.S. appeals court overturned a ruling that had temporarily barred Perplexity from using its AI-powered agentic shopping tools on Amazon's platform, clearing the way for continued use of their ML-based e-commerce assistant features.
Reddit Can Proceed With Scraping Claims Against Perplexity, SerpApi
yahoo.comReddit can proceed with a lawsuit alleging that Perplexity and SerpAi wrongly obtained copyrighted Reddit posts from Google's search results, challenging how AI search engines harvest data for their services.
Perplexity aims for the enterprise with AI-enabled browser, tools
tech.yahoo.comPerplexity launches new enterprise-focused AI tools including an AI-native browser. The move expands Perplexity's capabilities for business customers with specialized features tailored to corporate workflows and data needs.
Perplexity Returns with SPACE — Weekly AI Roundup
financialexpress.comPerplexity returns with SPACE, a new AI product or feature update alongside Chinese Kimi K3 model and OpenAI Codex hardware developments. The article discusses how Perplexity's system evolution focuses on smarter selection of models rather than just bigger ones.
Perplexity Brings AI Desktop Agent to Windows, Routing Tasks Across 20 Models
techtimes.comPerplexity expands its agentic desktop platform from macOS into native applications on Windows PCs, enabling users across nearly 1.4 billion devices while routing tasks between up to 20 different AI models. This extension broadens the availability of an important AI tooling solution for a large PC user base.
Reddit Can Proceed With Scraping Claims Against Perplexity, SerpApi
mediapost.comA judge rejected Perplexity's motion to dismiss the lawsuit accusing it and data-scraping services of vacuuming up Reddit content without permission, which is central to how AI search engines operate by pulling from Google's results. Article published on July 31, 2026.
AI search 'Perplexity' also to CLI, command line tools and agent collaboration
forest.watch.impress.co.jpDiscussion of Perplexity adding CLI capabilities for integration with agents.
Perplexity、Windows向けAIデスクトップエージェントを提供開始 20以上のモデルを自動使い分け
zaikei.co.jpPerplexity launched a Windows desktop AI agent that automatically utilizes over 20 models to handle tasks.
Perplexity Brings AI Desktop Agent to Windows, Routing Tasks Across 20 Models
techtimes.comPerplexity launched its Personal Computer for Windows on July 28, extending its agentic desktop platform from Mac to the operating system used by roughly 1.4 billion devices. The AI tool routes tasks across multiple large language models and integrates into local PC workflows.
Federal court clears Reddit in copyright suit against Perplexity AI
finance.yahoo.comFederal ruling allows Reddit to advance anti-circumvention claims against Perplexity AI and SerpApi in a high-stakes copyright case involving AI search data scraping.
Perplexity lance Personal Computer sur Windows
blogdumoderateur.comPerplexity expanded its AI personal assistant PC agent to Windows platform following a previous macOS deployment.
PerplexityがAIエージェントの"暴走"対策ツールをオープンソースに Claude CodeやCodexを監視
itmedia.co.jpPerplexity released Numbat, an open-source security tool to detect and prevent unsafe behavior in AI agents like Claude Code and Codex by monitoring their outputs.
Manhattan judge lets Reddit's DMCA claims against Perplexity move forward
msn.comManhattan judge refused to dismiss Reddit's copyright case against Perplexity AI, continuing legal challenges around data scraping for model training.
Perplexity says it plans to use Nvidia's new CPU
reuters.comPerplexity confirms plans to adopt Nvidia's next-generation CPU chips for its AI infrastructure, as the chip manufacturer broadens beyond GPUs. This strategic shift aims to optimize efficiency and reduce costs while maintaining high-performance inference capabilities in their distributed LLM training systems.
Reddit’s AI copyright lawsuit against Perplexity can move forward.
theverge.comFederal court allows Reddit's copyright lawsuit to proceed against Perplexity AI and data-scraping services SerpApi, accusing them of unauthorized scraping. The case centers on whether generative search platforms that aggregate web content infringe on rights without permission for their LLM training.
Germany strips AI search of its EU liability shield in world's first media ruling
msn.comGermany's media regulator ruled that Google AI Overviews and Perplexity outputs are subject to EU liability under new media law, marking world first formal application of this rule.
I gave Perplexity's agentic AI 5 complex tasks to run on my Mac - and I'll do it again
msn.comThe article documents a hands-on evaluation of Perplexity's agentic AI capabilities on macOS, testing the system with five complex tasks including web browsing, code execution, document analysis and cross-device coordination. The demonstration showcases advanced multi-step automation functionality that positions this conversational interface as a serious productivity tool for developers, researchers, and enterprise users requiring intelligent workflow orchestration across multiple devices and services.
Perplexity brings its Personal Computer AI agent to Windows
siliconangle.comPerplexity Inc. launched its Personal Computer AI agent for Windows, expanding its agentic automation software beyond macOS to bring the company's conversational interface directly onto the operating system running on most corporate and home PCs. The rollout represents a major step in making Perplexity's artificial intelligence capabilities more accessible across enterprise environments worldwide.
Perplexite's tokenmaxxing Model Council gives you multiple bot perspectives
theregister.comPerplexity expanded its Model Council to give multiple AI bot perspectives using token-maxing techniques on their Computer platform.