Anthropic Puts Claude in Slack as a Taggable Coworker
ALSO: Google sues over Gemini scams, world says China leads
Welcome to Guru99 AI Report!
Top Story: Today AI stopped feeling like a tool and started acting like a coworker — one you can now @mention in Slack. But that shift cuts both ways: as we hand machines more trust, the questions about jobs, security, and who’s really winning the race get louder. Let’s dig in.
🏷️ Claude Tag Turns AI Into a Slack Teammate
Brief Buzz:
Anthropic just blurred the line between coworker and software. On Tuesday it launched Claude Tag, embedding its flagship model directly into Slack as a taggable team member that remembers context, takes initiative, and works through tasks in the background while you focus elsewhere.
- Anyone in a channel can tag @Claude to delegate work; one shared Claude serves the whole team, so colleagues can pick up where others left off.
- It builds persistent memory from channel history — no re-explaining context every time.
- An optional “ambient” mode lets Claude jump in unprompted to flag updates and chase stalled threads.
- Anthropic says 65% of its product team’s code is now written by an internal version of the tool.
- Available in beta for Enterprise and Team customers, with other platforms coming next.
💡 Why Should You Care?
Treating AI as a colleague means trusting it more — raising the stakes on accuracy and data security, and stoking fears that it replaces jobs rather than augmenting them.
🎣 Google Sues the Scammers Who Weaponized Its Gemini AI
Brief Buzz:
Those “package held” and “overdue toll” texts? Google just sued the operation behind them. Outsider Enterprise, a China-based network, allegedly used Gemini to mass-produce phishing sites and scam texts — the first time Google has sued anyone for weaponizing its own AI.
- 2.5 million scam texts hit Android users in one two-week May stretch — 55,000 spam complaints.
- Court filings cite 3.87 million stolen cards and roughly $1.9 billion in losses since July 2023.
- Gemini generated the code for 9,000+ fake sites spoofing Google, the USPS, banks, and toll agencies.
- The kit sold like a franchise: $88/week on Telegram, 290+ templates, no coding required.
- Google is now coordinating with the FBI and carriers, and backing seven bipartisan anti-scam bills.
💡 Why Should You Care?
AI collapsed the cost of convincing fraud. Building a fake bank site once needed a coder and hours — now it needs $88 and a Telegram login. Read more in Google’s breakdown of the takedown.
🤖 GM’s New Cobots Anger Union After 1,000+ Layoffs
Brief Buzz:
General Motors quietly rolled out about 50 collaborative robots at its Detroit Factory Zero plant — the same site where more than 1,000 workers were laid off as EV demand cooled. The timing has the United Auto Workers fuming and previewing a fight over automation.
- ~50 Fanuc cobots now attach body panels on the line, working cage-free beside the employees who remain — not behind safety barriers.
- UAW Local 22 says members are “disgusted” and has filed grievances over both job losses and safety.
- GM frames it as a safety and ergonomics upgrade, and CEO Mary Barra has cast AI as letting workers “focus on craftsmanship.”
- The layoffs trace to slumping EV demand after the $7,500 federal tax credit ended — and the plant already ran 1,000+ robots, so 50 cobots aren’t a one-for-one swap.
- GM isn’t alone: Toyota put humanoid Digit robots in its Canadian RAV4 plant, and BMW is extending humanoid trials from South Carolina to Germany.
💡 Why Should You Care?
This is a live test of whether AI and robots augment factory workers or quietly erase them. With UAW’s 2028 contract looming, Factory Zero is becoming the bellwether for blue-collar automation.
🧬 Meet the AI Lab Decoding the Grammar of DNA
Brief Buzz:
We usually think of AI as built on language. Radical Numerics is building it on the “grammar” of DNA. The San Francisco lab emerged from stealth with $50M to pursue general biological intelligence — AI that reads, writes, and designs across all of biology.
- Its founding team built the open-source Evo and Evo 2 models — the first AI that can both read and write DNA at scale — later used to create the first complete AI-designed genome, a bacteriophage.
- New model Omnii, a “genome language model,” is in research preview — but unlike Evo, it’s proprietary, letting the lab control who gets access.
- That reflects a dual-use dilemma: tech that could cure disease could also help build dangerous pathogens. The lab pitches itself as doing both biological design and defense.
- It joins a crowded race — OpenAI’s GPT-Rosalind, Anthropic’s Claude for Life Sciences, and DeepMind’s Nobel-winning AlphaFold all aim AI at the life sciences.
💡 Why Should You Care?
AI is shifting from writing text to writing biology. That could accelerate cures for diseases like cancer — but the same power to design DNA on demand is why controlling access suddenly matters so much.
🎬 AI Filmmaking Comes of Age — and Hollywood’s Betting Big
Brief Buzz:
Runway’s fourth annual AI Film Festival just screened ten AI-generated shorts pulled from thousands of submissions — and the leap in quality has Hollywood paying attention. From studio equity deals to A-list directors, the industry’s relationship with generative AI is shifting from skepticism to strategy.
- Telltale glitches remain — off-sync lips, too-smooth skin, garbled on-screen text — but the work has come a long way from those viral “Will Smith eating spaghetti” clips.
- Lionsgate deepened its Runway bet, taking an equity stake and launching a joint program to co-develop new IP with generative models.
- Martin Scorsese signed on as an adviser to image-AI startup Black Forest Labs, using its tools to storyboard his next film.
- Backers pitch AI as a way to “raise the floor” — letting first-time filmmakers get long-stalled projects off the ground — even as critics fear studios will use it to cut creative jobs.
💡 Why Should You Care?
Making a movie is getting radically cheaper. That could unleash more creator-driven hits like Backrooms and Obsession — while raising hard questions about craft, jobs, and what virtuosity means when anyone can create.
🐉 The World Now Thinks China Is Winning AI
Brief Buzz:
The US built the AI boom — but the world increasingly thinks China is winning it. A 15-country Public First poll found 11 nations now see China as the AI frontrunner, even as American optimism about the technology keeps sinking.
- In 11 of 15 countries — including allies France, the UK and Canada — most now see China leading; only the US, Japan, India and Vietnam still back America.
- Just 51% of Americans think the US is ahead, versus 24% who say China.
- US optimism that AI will improve society slid from 39% (2024) to 31% (2026), and Pew finds 50% of adults more concerned than excited.
- 7 in 10 Americans now oppose new data centers nearby, citing water, energy and quality-of-life worries.
- In China, under 10% fear AI will take their jobs — and Stanford says the US–China model gap has nearly closed.
💡 Why Should You Care?
Perception shapes markets. If buyers, developers and allies believe China builds the best — and cheapest — AI, that mindshare pulls talent, investment and adoption its way, no matter who’s actually ahead on the benchmarks. Read the full global perception findings.
Hey! I’m Krishna Rungta
Founder of Guru99.com, Editor-in-chief & Technology Expert
Was this email forwarded to you? Sign up for free here.
