We're all using AI now fine. But PLEASE can we at least read what the AI wrote before sending? Maybe even make a passing attempt to make it sound 5% human?
You're still saving 90% of the time and you don't look like someone who has outsourced the entirety of your cognitive processes.
1
26
Exchanges are giant honey pots. One word.
SELF-CUSTODY
BREAKING: World's largest crypto exchange Binance hacked with over $5.6 billion stolen:
• $1.45 billion $BTC
• $875 million $ETH
• $234 million $TRX
• $50 million $XAUT
• $1.57 billion $USDT
• $823 million $USDC
• $75 million $WBETH
• $88.5 million $RLUSD
+ $447.5 million additional cryptocurrencies
This is the largest hack in crypto history.
1
50
Chris Georgen retweeted
Crypto advocates are putting yard signs up across the country to show that they are voting for crypto on November 3.
Crypto voters are everywhere.
If there's one thing that AI has supercharged, it's phishing scams and other attempted cons and exploits targeted at startup founders (and probably everyone else as well).
I've cleared two things off of my calendar this week already that were clearly fraudulent.
3
3
51
These groups use AI builders to create authentic looking websites and bots to manufacture enough social media activity to look legitimate. Thankfully there are still tells if you know where to look.
The latest example I found lists a few phone numbers on their website (London, NYC, Singapore, very impressive looking). However, a quick google search shows that one of the numbers actually belongs to Brevan Howard and was just lazily hallucinated by whatever AI was used to make the website.
1
1
22
It's honestly getting more and more difficult to spot these sorts of scams, but thankfully your friendly neighborhood AI agent is just as good at helping you as it is at helping the scammers.
When in doubt, just take whatever info you have and pass it off to an agent to do a bit of research. It's not foolproof, but it's a very strong start.
Stay safe out there.
12
Chris Georgen retweeted
The only thing that stops a bad guy with an open weight model is a good guy with an open weight model
Le Chonk. Amazing example of leaning into the meme 😂
Meet Mistral Large 4, aka Le Chonk.
• 1T parameters, natively multimodal. 49B active.
It is the best open weights model from US or Europe on aggregated benchmarks.
• State-of-the-art on critical workloads, including cyber defense, manufacturing and finance and it surpasses closed frontier models on visual grounding.
• Forged in Europe end-to-end and is deployable from Europe via our own Mistral Cloud infrastructure.
• Available to all via API today. Working with cybersecurity partners privately.
Open weights release end of October.
94
Right now tokenization is kinda boring. Invest in the same things you've always been able to with now with a higher transaction fee.
MetaLex is different. MetaLex is using tokenization to make early stage venture investing more transparent, dynamic, and more accessible.
We're excited to be collaborating with them to bring their amazing venture offerings to Ampersand users through our open investment marketplace.
There’s a lot of activity around AI agents for investing.
We think the more interesting question is what happens when you’re not limited to one agent, one strategy, or one financial product.
This is where Ampersand comes in.
2
3
125
B2C, < 30 years old : TikTok, Snap
B2C, 30-45 years old: Instagram
B2C, 45 years +: Facebook
B2B: LinkedIn
X is just for the first 20 free trials users you get when your product doesn't work yet. They don't convert to paying later but damn do they sound excited.
2
3
199
1 // What's happening on Fomo, Pump and rest of the memecoin space is absolutely fascinating right now.
There are some pretty novel ideas around how to structure stock or RWA rewards into tokens as well as how engagement and community can be monetized.
2 // Ideas around how to activate super engaged communities around sports or even Pokémon cards.
I saw one project where creator fees were being used to pay for long running agents that were meant to evolve in personality and knowledge as they gained more access to inference--felt like the agentic version of CryptoKitties.
1
1
64
3 // All that said, everything there is a literal rug pull. Ideas that legitimately could have an audience and could conceivably be built in only a few weeks are actually just being used as scams to farm a couple thousand dollars in trading fees.
So if you're looking for new financial primitives or community-driven startup ideas, head over to the dark underbelly of crypto for inspiration.
1
30
Illinois crypto advocates in have reached out to senate offices more than 40,000 times in support of CLARITY
@SenatorDurbin @SenDuckworth, we've shown up.
Today, make our voices count. Vote YES to advance CLARITY. We're watching, and we vote.
AI is not going to do anything on its own, and iterative improvements are not going to change that. The real and quite probable risk is SOMEONE will use AI as a weapon. It will undoubtedly be a very effective weapon too.
This is the way we should be framing the conversation, but unfortunately believing that AI will very soon gain agency and not simply remain a tool doesn't speak to the god complexes of the people who believe they will birth AGI.
Jacob is correct here—we really do earnestly believe AI could kill all humans! I personally think it is >10% within the next decade. I believe Anthropic is trying its best, but we do not yet have a plan to solve alignment for superintelligence and are not clearly on track to.
1
3
171












