Get Even More Visitors To Your Blog, Upgrade To A Business Listing >>

When AI bots impersonate humans


Welcome to Startup Weekly. Register here to receive it in your inbox every Friday afternoon.

This week, I’ve been thinking a lot about the implications of artificial intelligence. One of the most fragile parts of the puzzle is the training data. We already know that you can see if your images were used to train the data sets and that many of the training data sets that exist are . . . spurious at best. Some startups are trying to create data sets trained exclusively on licensed dataand human artists are quite grumpy when Renowned studios use AI to generate art.

One interesting curve to realize is that even evaluating training data can be challenging, as researchers find that Mechanical Turk workers, who are, in theory, human workers performing tasks that machines cannot, supposedly They are using artificial intelligence tools. That’s fine for some tasks, but not great if the text you’re generating is intended to be used as the rubric against which AI-generating text tools are measured. The old computer adage of “garbage in, garbage out” still stands. If you cannot trust the training data, you can’t trust the output (TC+).

Viewed through the lens of startups, AI continues to triumph, and Amazon’s AWS is throwing its fair share of weight behind the burgeoning trend, with a $100 million program to finance generative AI initiatives.

The highest highs and deepest lows of the transport

Image Credits: Bryce Durbin/TechCrunch

It’s a bad week when a submersible goes missing, followed by a spate of reports showing the company may not have the strongest safety record. The submarine imploded under the crushing pressure of the ocean, killing all five of its passengers and raising new questions about the role of whistleblowers on the startup land. The OceanGate underwater vessel used a carbon fiber hull that “was not rated for the depths of the Titanic”said the director of operations turned whistleblower. This goes to show that while startups are famous for “moving fast and breaking things,” that principle may not fully extend to life-or-death critical equipment.

The highlight for transportation this week is a breakthrough in battery manufacturing. It turns out that up to half of the energy required to make a lithium-ion cell is used in the drying process of certain components of the battery cell. Volkswagen has just introduced a new process (TC+) which eliminates that requirement, drastically reducing the cost and time it takes to create the batteries that power our electric vehicles.

Hackers are going to hack

Image Credits: Bryce Durbin/TechCrunch

In both the startup and non-startup arenas, we’re seeing a tremendous amount of movement in security news in recent weeks. We’ve had a lot of coverage from Reddit slowly imploding about the new API charges that the social media giant introduced a while back. But one story that I particularly wanted to highlight is that hackers threaten to release confidential data stolen from Reddit unless the company pays a ransom demand, and reverses its controversial API price increases. It seems quite curious that hackers demand both policy changes and cash. If they were given one or the other, I wonder which one they would choose, and what does that say about the power of hacktivism?

Malicious hacking has been in the news for a long time, but I find it particularly interesting that we are seeing more and more startups trying to help address the problem, either API security hardening, data security at source either Internet of Things Devices. As hackers become more sophisticated and scanning for computer security exploits becomes more frequent (just last week, a gang of ransomware listed its first victims of the massive MOVEit hacks, which included US banks and universities), opportunities for startups are also increasing. And yet, as Alex recently explored, it seems like a bug that Venture Capitalists Don’t Line Up Around the Block (TC+) to finance the current generation of cybersecurity companies.

You’re so money and you don’t even know it

Image Credits: Getty Images/Svetlana Borovkova

Fintech, what are we going to do with you? Even in an industry that is all about money, the vertical continues to explode with investment. Paro raised $25 million to match independent financial experts with businessesopen banking fintech company Volt just lifted a big round with a valuation north of $350 million, and Majority, a digital bank for American immigrants, closed almost $10 million as it expands its operations in Texas.

It’s not sexy, but investors know that fintech, once the company gets a lot of traction and a handful of customers, is interesting for two reasons: money never goes out of style, and there are multinational giants hungry for M&A. who are standing to take over a booming company. At startup scale, the most recent example was that Robinhood just acquired credit card startup X1 for $95 million. In fact, on a very different scale, Nasdaq announced that it plans to acquire financial services software company Adenza for $10.5 billion..

Looks like there’s money in those money trees.

  • Wait, who am I paying?: Catherine reports that Notarize Launches Test and doubles down to ensure secure(r) online transactions.
  • That’s it, no more scams for you.: Mary Ann reports that Plaid reveals a new collaboration network intended to share fraud intelligence.

Top reads at TechCrunch this week

Image Credits: Jose Giacomin (Opens in a new window) / Fake Images

If you’re having a hard time raising money, there’s probably one of three reasons investors remain indifferent: the market may be too small, the team isn’t good enough, or your plans just don’t make sense. In my latest article on TechCrunch+, I discuss how venture capitalists evaluate those thingsand how you can push back.

If you can’t beat them with a tablet…: You can think outside the box. Brian discusses the new Google Pixel tablet and concludes that it’s all about the pier. You will have your work cut against Apple’s iPad Pro-looking entry-level iPads.

Welcome to Streamberry: Netflix’s doppelgänger is personalized for your worst nightmares. Lauren reports that Netflix launches a website based on the fictional “Black Mirror” streaming service. In that sense, don’t miss the article I wrote about how technology unlocked Netflix’s business model (TC+).

Leaping into the quantum realm: Frederic reports that Microsoft expects build a quantum supercomputer In 10 years. Maybe that’s why they need that fusion reactor?


Get your TechCrunch IRL fix. Join us at Disrupt 2023 in San Francisco this September to dive into all things startup. From headline interviews to intimate panel discussions to a packed startup show floor, there’s something for everyone at Disrupt. Save up to $600 when you purchase your pass now through August 11, plus save 15% on top with promo code STARTUPS. Learn more.



Source link

The post When AI Bots Impersonate Humans appeared first on Interview Preparation.



This post first appeared on Interview Preparation, please read the originial post: here

Share the post

When AI bots impersonate humans

×

Subscribe to Interview Preparation

Get updates delivered right to your inbox!

Thank you for your subscription

×