I've verified from reliable sources. New visual Turing Test: this is a real robot demo! The controller is a neural net trained in Isaac simulator using reinforcement learning and then sim2real. Reward engineering is all you need.
Walking gait's got swag but we need these robots to go fire-fighting asap!
Below is a video shot by another passer-by from a different angle. There's even a pic of the robot falling and engineers rescuing it (in thread):
Notice how logos recently all look the same?
Not because it makes them look better...
But because of THIS psychological trick that manipulates your brain.
That's why Google, Microsoft, and Airbnb are all doing it.
Here's the full explanation:🧵
This week’s episode of #LIDT, entrepreneur Alex Wright tells us, how thinking differently helped to create @DashDrinks. Find out:
🔥How they won 57% of the market
🏆How failure led to success
✍️ 3 secrets to starting a business
Watch the full episode: https://t.co/rlM2hEvEhw
Famously Facebook turned down Yahoo's $1B acquisition offer in 2006 and is now worth $1.28T.
For many it does not turn out well:
▫️ Groupon declined a $6 billion offer from Google in 2010. Since then, its valuation has dropped to $600M
▫️ Moz declined a $25M M&A deal from Hubspot in cash and Hubspot stocks. In 2014, HubSpot had one of the most successful IPOs of any SaaS company.
▫️ Path rejected a $100 million buyout from Google. It was eventually sold for pennies.
▫️ Yahoo! turned down a $44.6 billion offer from Microsoft in 2008. It later sold to Verizon for just $4.8 billion in 2017 after years of declining ad revenue and strategic missteps.
▫️ Digg declined a $200 million offer from Google in 2008. It struggled and was eventually sold in parts for about $500k in 2012.
▫️ Qwiki turned down a $150M offer from Google and was later sold for $50M.
▫️ Friendster rejected a $30M offer from Google in 2003. The platform failed to scale effectively and couldn't keep up with Facebook. Eventually, it transitioned into a gaming site before being shut down in 2018.
▫️ Viddy turned down a $100M offer from Twitter in 2012 and was eventually sold for $15M in 2014.
One of the most under-the-radar contrarian idea Peter Thiel ever articulated was that failures teach you nothing.
Just think of how much mythology exists around justifying failures as “oh well I learnt this and that”. It’s practically a PMC shibboleth to “learn from your failures and mistakes” all with rote interview questions surrounding “so tell me about a time you failed”
And yet, Thiel makes the claim that you don’t learn anything useful from failures — because failure is always overdetermined.
In other words, if something doesn’t work, it could have also *not worked* for 99 other reasons.
Over the last year, the more things go right at my startup, the more I realize how true Thiel’s original point was.
Each second, 1.5 million tons of solar material, traveling at 100 miles per second, shoot off the sun. Earth's magnetic field deflects most of it, but not all. The solar wind, a stream of charged particles, flows at 447 km/s (1 million mph), and while the magnetic field protects us, some particles still cause auroras and geomagnetic storms.
I will trust you – I will extend my hand to you – despite the risk of betrayal, because it is possible, through trust, to bring out the best in you, and perhaps in me.
Sama inserts the secret recipe into a GPT: "I'm going to say I want to help startup founders think through their business ideas and get advice. After the founder has gotten some advice, grill them on why they are not growing faster."