Joined December 2017
- Tweets 34,732
- Following 633
- Followers 46,576
- Likes 173,049
4,019 Photos and videos
Pinned Tweet
37
308
2,348
1,576,128
Jun 13
🚨 News: Google is suing a Chinese cybercrime group it says used Gemini to run financial scams against hundreds of thousands of Americans.
the group, called Outsider Enterprise, used gemini to build hundreds of fake websites posing as companies like google and youtube, plus government services like the Postal Service and New York's E-ZPass toll system.
google says it's the first time it's coordinating with the FBI and carriers like AT&T, T-Mobile, and Verizon to take a network down.
the scale is large. the group built 131 software kits to mass-produce fake sites.
In just two weeks in May, it sent 2.5 million messages to Android users linking to 9,000 fake websites.
the FBI says AI-driven fraud is growing faster than other scams.
of the roughly $21 billion lost to fraud last year, about $893 million was tied to AI.
4
13
21
4,368
Jun 13
open source AI labs satisfaction level rn.
Jun 13
The US government, citing national security authorities, has issued an export control directive to suspend all access to Fable 5 and Mythos 5 by any foreign national, whether inside or outside the United States, including foreign national Anthropic employees.
The net effect of this order is that we must abruptly disable Fable 5 and Mythos 5 for all our customers to ensure compliance.
Access to all other Claude models is not affected.
We apologize for this disruption to our customers. We believe this is a misunderstanding and are working to restore access as soon as possible.
Read our full statement: anthropic.com/news/fable-myt…
2
9
17
1,649
Jun 13
“i don’t ever give up. i mean, i’d have to be dead or completely incapacitated”
Jun 12
“It is certainly hard to believe that a little company that started in a warehouse in El Segundo is now going public with the largest IPO ever.”
Only in America. Congratulations to @elonmusk @SpaceX. This is the American Dream 🇺🇸💪🚀
10
21
2,100
Jun 12
News: Three trillion-dollar tech IPOs are landing at once, and they could create around 20 new billionaires.
SpaceX raised about $75 billion this week at a $1.77 trillion valuation.
anthropic the maker of Claude and OpenAI have both filed to go public soon, each valued close to $1 trillion.
a markets research firm, Sacra estimates the three listings could create over 20 billionaires from current and former staff plus more than 16,000 millionaires.
that list includes names like Greg Brockman who testified his OpenAI stake was worth $30 billion and Dario Amodei, already estimated at around $8 billion.
part of why the numbers run so high, these companies stayed private for years so equity piled up before hitting the public market.
ALT source: https://www.nytimes.com/2026/06/12/business/economy/spacex-openai-anthropic-billionaires-ipos.html
4
9
29
4,670
Jun 12
🚨News: SpaceX just made history with the largest iPO ever, and Elon musk has become the world's first trillionaire
here's what actually happened that you should know 👇
ipo price: $135 a share, a fixed price spacex set itself.
shares sold: 555,555,555 (all newly issued).
capital raised: ~$75 billion.
valuation: ~$1.77 trillion.
that's about 3x the previous record (saudi aramco, 2019).
→ day one
opened at $150, hit a high of $176.52, finished up ~28%.
market cap pushed past $2.2 trillion, bigger than meta and samsung.
360M shares traded by early afternoon, about 10x the next-biggest ipo of the year.
elon musk became the world's first trillionaire on paper (~$1.1T).
→ what's actually inside the company
it went public as one entity, three segments:
space (falcon, dragon, starship)
connectivity (starlink)
ai (xai / grok / x, after the feb 2026 merger)
starlink wasn't spun out, it's the core engine.
→ the starlink numbers most people don't know
$11.4B revenue in 2025, 61% of all spacex revenue.
grew from 2.3M customers in 2023 to 10.3M by march 2026.
~9,600 satellites in orbit, roughly 75% of all active maneuverable satellites on earth.
→ the launch record nobody talks about
spacex flew 165 falcon 9 orbital missions in 2025.
that's more than the rest of the world combined.
sixth year in a row it set a new launch record.
→ the demand was staggering
order book hit ~$250 billion, 3.5 to 4x oversubscribed.
blackrock alone reportedly ordered $5 billion.
retail got an unusually large slice (~20% , vs. the typical 5 to 10%).
retail orders alone topped $100 billion.
→ what the money is funding
finishing starship development.
expanding starlink (musk cited plans for 100,000 satellites).
building data centers, including orbital ai compute, "data centers in space," targeted as soon as 2028.
→ a few records it set in one day
largest ipo in history.
became roughly the 6th to 8th most valuable u.s. company instantly.
created the world's first trillionaire.
underwriters even wore green shoes to the listing, a nod to the "greenshoe" option.
this was one of the biggest market events in modern history.
1
2
13
4,402
Meer | AI Tools & News retweeted
Jun 10
🚨News: Google just dropped DiffusionGemma, an open model that writes text the way image generators draw pictures all at once instead of word by word.
most language models build a sentence one token at a time, like a typewriter.
DiffusionGemma drafts an entire 256-token block in parallel, then refines it over several passes until it converges.
the payoff is speed: up to 4x faster on dedicated GPUs, with 1000 tokens/sec on an H100 and 700 on an RTX 5090.
it's a 26B Mixture of Experts model that activates just 3.8B parameters fitting under 18GB VRAM when quantized.
bi-directional attention also helps with code infilling and in-line editing.
Google says output quality runs lower than standard Gemma 4 so it's aimed at speed-critical local work not max-quality production.
Weights are out now on Hugging Face under Apache 2.0.
Meet DiffusionGemma ⚡ Our latest experimental open model (Apache 2.0) that generates text up to 4x faster.
Instead of predicting and typing just one word at a time like most language models, it drafts and refines entire blocks of text simultaneously.
Here’s how it works 🧵 ↓
8
13
52
9,437
Meer | AI Tools & News retweeted
Jun 10
🚨News: Meta is building its first AI data center in India, and it's teaming up with Reliance to do it.
the project is a 168-megawatt AI-enabled facility in Jamnagar, Gujarat.
Meta is leasing the capacity while Reliance handles everything from design and construction to power, connectivity and day-to-day operations.
the center will run on renewable energy and use desalinated seawater for cooling.
Meta is covering the full cost of the energy and water it uses.
Reliance says it'll be ready within two years and can expand over time.
It will also plug into Meta's global AI computing network.
On top of that, Meta has contracted nearly 1 gigawatt of new renewable energy in India through CleanMax and Fourth Partner Energy.
the deal builds on Meta's $5.7B investment in Jio back in 2020. Value of the new agreement wasn't disclosed.
ALT source: https://techcrunch.com/2026/06/10/meta-signs-first-ai-data-center-deal-in-india-with-reliance/
9
7
26
31,013
May 25
no way!!! …..
this shouldn’t feel this easy
if you run an online store, you need to see this.
GPT Image 2.0 Seedance 2.0 in Pollo AI did something I didn’t expect at all.
it changed how content gets made 👇
7
4
17
12,231
May 25
I tried the AI Avatar feature and created a Bleu De Chanel-style ad with Cristiano Ronaldo as the model.
The result was mind blowing.
I didn't expect that.
1
1
529
May 25
After using @itsPolloAI for a bit, here’s the honest take:
It won’t replace everything.
But it removes a lot of the “I’ll do this later” delay.
And that’s where most brands get stuck.
Try it yourself 👇
Web：tinyurl.com/3dv5nznb
App：pollo.ai/download
397
May 22
the field of agentic AI moved faster than its job titles
most of the senior people doing this work don't show up cleanly on linkedin
ran this query through @nimble_search talent sourcing skill:
> find senior AI engineers in SF Bay Area with 5 years in agentic systems or multi-agent frameworks, strong Python, recent startup experience
the agent browsed the open web and read what each person has actually built
compound constraints like this break keyword search
reading pages like a human is what turns a query like this into an actual shortlist
6
16
26
12,447
🚨 A $2.5B startup just put Nvidia in a sandwich on the hardest document retrieval benchmark in AI.
It's called webAI-ColVec1. And they open sourced it.
Their 9B model sits at #1 on ViDoRe V3. Their 4B model sits at #3. Nvidia's best open-source embedding model is stuck at #2 between them.
ViDoRe V3 is not a toy benchmark. 26,000 document pages. 3,000 human-verified queries. 10 enterprise domains.
Financial filings, healthcare records, technical manuals, dense tables, messy layouts. The stuff that actually breaks production RAG systems.
Here's what makes this different from everything else on the leaderboard:
→ Retrieves directly from rendered page images instead of extracted text
→ Skips OCR entirely. The model sees the page the same way you do
→ Tables, charts, scanned pages, dense layouts. All handled natively
→ Two model sizes: 4B for speed-sensitive edge deployments, 9B for max accuracy
→ Trained on ~2 million question-image pairs across scientific papers, financial filings, government reports, healthcare docs, and multilingual documents
→ Built on Qwen 3.5 vision-language backbones with LoRA adaptation
→ Trained on just 8 A100s with an effective batch size of 512
→ Each query learns against 511 competing document pages per training step
→ Proprietary loss function that forces cleaner
separation between correct and wrong pages
→ Multiple embedding sizes (128, 640, 2560) so you pick your own speed vs. quality tradeoff
Here's the wildest part:
Most enterprise teams are paying per-page and per-token fees just to get their documents into a format their RAG system can search.
Reducto charges $0.015 per page for parsing. Cohere Embed v4 costs $0.12 per million tokens.
Voyage AI's flagship model runs $0.18 per million tokens.
And all of those still depend on OCR as the first step. One bad table extraction upstream and your entire retrieval pipeline breaks.
webAI threw out that entire architecture. The model reads the page like a human.
And it beats every paid and open-source alternative on the benchmark designed to test exactly that.
This didn't come from a massive model or a giant infrastructure budget. 8 A100s.
Deliberate training recipe. Retrieval-specific design. That's it.
Cohere Embed v4: $0.12/million tokens. Voyage AI voyage-3-large: $0.18/million tokens.
OpenAI text-embedding-3-large: $0.13/million tokens.
This: Free. Open source. #1 on the leaderboard.
@thewebAI
100% Open Source.
(Link in the comments)
11
6
19
21,341
Meer | AI Tools & News retweeted
May 1
Your Ad Flow just got an upgrade with Ad Flow Copilot.
Idea in. Storyboard in chat. Brand assets in. Ad out.
or skip ahead: "Copilot, make the ad"
Start here:
44
52
259
249,531