Top Tweets for #aithisweek
𧡠#AIThisWeek: What if the next big leap in AI doesn't come from a new model? π€
Google is improving the systems around agents, Perplexity is strengthening its security, Anthropic is making them proactive, and Inception and MiniMax are making them faster and cheaper. Meanwhile, OpenAI is advancing legacy code modernization, GitHub is rethinking developer workflows, and Krea is prioritizing AI-powered image editing. π
𧡠#AIThisWeek: Mistral and Reflection AI are pushing powerful open-weight models, while Google is taking a more controlled approach with Gemini 4 Argon, giving cyber defenders access first. At the same time, Claude is moving deeper into government, Google Docs is adapting to AI-generated workflows, OpenAI is bringing visual ads to ChatGPT, and Pinterest is turning saved inspiration into action.π
𧡠#AIThisWeek: Agents took another step into commerce, infrastructure, devices, and research this week. OpenAI launched Dots to take on Meta's Muse, Shopify opened checkout to agents, and Automattic launched infrastructure for AI-built projects. At the same time, OpenAI's safety issues showed the risks that come with giving agents more autonomy. π
𧡠#AIThisWeek: Google is building Gemini directly into the laptop OS. Snap is putting an AI assistant behind AR glasses. Meta is opening Muse to third-party services, while Google's CC is becoming a shared agent for entire households. Anthropic and OpenAI both put out cheaper flagship-class models the same day. And inside Anthropic, Claude is already leading 26% of the company's AI R&D work. π
𧡠#AIThisWeek: Agents are moving everywhere. OpenAI is preparing managed agents. Meta launched one that can spend money for you. An Anthropic researcher quit over the race toward self-improving AI. And thatβs only part of what changed this week. π
𧡠#AIThisWeek: What happens after the model works? OpenAI shipped GPT-6 Astra with access off by default. Anthropic split Fable and Mythos 5.1 by who is allowed to use them. Runway is generating interfaces instead of coding them. Cohere is competing on economics. And Cursor lost OpenAI access because someone bought it. π
𧡠#AIThisWeek: Faster intelligence is bringing bigger consequences.
Google shipped Gemini 3.7 Flash just weeks after its predecessor; OpenAI previewed 14x-faster GPT-5.6 Sol; https://t.co/9SLdJ7C0yP delayed GLM-5.3's weights after unexpected cyber capability gains; and Cursor launched a GitHub rival during a six-hour GitHub outage. π
𧡠#AIThisWeek: Agents are moving into the parts of the web that were built for people.
Meta released an agent model that runs on a consumer GPU, Cloudflare built a browser specifically for AI agents, and Google Maps started placing food orders.
Meanwhile, Adobe moved 70+ creative tools into ChatGPT, and Cloudflare forecast a web where machine traffic could outnumber human traffic 1,000 to one. π
𧡠#AIThisWeek: MCP is going stateless, WordPress VIP introduced governed agent access, OpenAI released Presence for production deployments, Perplexity rebuilt the sandbox layer, Moonshot challenged the frontier, Adobe tracked creator adoption, and more. π
𧡠#AIThisWeek: Anthropic launched Claude Sonnet 5, introduced Claude Science for researchers, restored Fable 5 after US export controls were lifted, and proposed a new industry framework for evaluating AI jailbreaks.
Meanwhile, OpenAI previewed GPT-5.6 behind a government access gate, Codex became the company's default AI tool, and autonomous agents continued expanding into everyday work. π
𧡠#AIThisWeek: Anthropic faced the first government-mandated recall of a frontier model, OpenAI and 1Password made major bets on agent governance, and the EU moved closer to mandatory AI transparency rules. Meanwhile, Midjourney unveiled a full-body medical scanner in a surprise pivot from image generation. π§π
𧡠#AIThisWeek: OpenAI moved into personal finance. Anthropic acquired a shared piece of developer infrastructure and cut competitors off from it. KPMG deployed Claude across 276,000 employees. Cohere made a major push for sovereign AI. And coding agents from OpenAI, Cursor, and xAI kept moving closer to autonomous software work. π
𧡠#AIThisWeek: When AI stops waiting for prompts, the real question becomes control. Agents are now triggering workflows, crossing system boundaries, and acting without being asked. At the same time, the race is shifting from models to implementation. π
π§΅#AIThisWeek: OpenAI, Anthropic, Canva, Adobe, and Salesforce all move in the same direction: the platforms teams already use are being rebuilt around agents. Perplexity pushes into the operating system layer. Canada opens applications for a $890M national AI supercomputer. π
𧡠#AIThisWeek: Meta launches Muse Spark and jumps from No. 57 to No. 5 on the App Store overnight. HumanX puts 6,500 enterprise leaders in a room and the conversation keeps coming back to one word: agents. Microsoft confirms it's building an enterprise-grade answer to OpenClaw. MontrΓ©al's Hypertec becomes Nvidia's first OEM partner in Canada. LinkedIn wants to be the platform that connects professionals with AI training work. And Stanford's AI Index finds acceleration outpacing governance by a wide margin. π
#AIThisWeek: Anthropicβs accidental leaks reveal a new model and an autonomous agent roadmap, Mila and Mozilla partner on a portable AI memory layer, Bluesky launches Attie for user-controlled feeds, Runway expands into ecosystem-building with a $10M fund, and OpenAI shuts down Sora after high costs and declining usage.
π https://t.co/5JYMa8q0rr
π§ Prefer listening?
Each AI This Week post includes an audio overview embedded directly in the post for easy listening.
#AI #News #Tech #EnterpriseAI
𧡠#AIThisWeek: OpenAI outlines its roadmap to autonomous AI researchers, Anthropic gives Claude computer control, Cisco benchmarks LLM security, Perplexity moves into health data, Adobe expands Firefly, Cohere launches top-performing speech recognition model, Google improves model efficiency, and Apple sets the stage for WWDC26. π
Thatβs a wrap.
For the full breakdown and analysis, explore the latest edition of AI This Week.
π https://t.co/jVZxKgFUMp
#AI #AIThisWeek #GenAI #EnterpriseAI #AIPlatforms #WordPress #Google #Nvidia
π§΅#AIThisWeek: OpenAI released GPT-5.4 and acquired a security platform, Anthropic launched a marketplace and multi-agent code review, Google embedded Gemini deeper into Workspace, Adobe expanded conversational editing, and a16z reframed the competition as a platform war. π
π§΅#AIThisWeek: Frontier reasoning models improved, multilingual AI shrank to run locally, agents moved into messaging apps, and security became a product feature. Capability accelerated, but so did the risk conversation. π
Last Seen Hashtags on Sotwe
minichat(*)**(***)****(*****)******(********)
Seen from Turkey
anxiouspanda()+filter:native_video
Seen from United States
ryma
ensesttΓΌrbanlΔ±
Seen from Turkey
japanese fart
Seen from Turkey
DisventureCamp
Seen from Colombia
minichat()(()*(()*()**))+filter:videos
Seen from Mexico
Ψ³ΩΨ¨Ψ±ΩΨ§ΫΩ
Seen from France
teenageers
Seen from Turkey
lorengray
Seen from Colombia
Most Popular Users
Elon Musk 
@elonmusk
241.5M followers
Barack Obama
@barackobama
118.8M followers
Cristiano Ronaldo
@cristiano
114.5M followers
Donald J. Trump
@realdonaldtrump
111.8M followers
Narendra Modi
@narendramodi
107.1M followers
Rihanna
@rihanna
98.7M followers
NASA
@nasa
92.3M followers
Justin Bieber
@justinbieber
91.8M followers
KATY PERRY
@katyperry
90M followers
Taylor Swift
@taylorswift13
84M followers
Lady Gaga
@ladygaga
75.5M followers
Virat Kohli
@imvkohli
73.5M followers
Kim Kardashian
@kimkardashian
70.9M followers
YouTube
@youtube
68.8M followers
Neymar Jr
@neymarjr
66.5M followers
Bill Gates
@billgates
65.3M followers
Selena Gomez
@selenagomez
63.1M followers
The Ellen Show
@theellenshow
62.2M followers
CNN
@cnn
61.7M followers
X
@x
60.7M followers