Latest news with #TechMemo
Business Insider
2 days ago
- Business
- Business Insider
OpenAI and Microsoft are dueling over AGI. These real-world tests will prove when AI is really better than humans.
AGI is a pretty silly debate. It's only really important in one way: It governs how the world's most important AI partnership will change in the coming months. That's the deal between OpenAI and Microsoft. This is the situation right now: Until OpenAI achieves Artificial General Intelligence — where AI capabilities surpass those of humans — Microsoft gets a lot of valuable technological and financial benefits from the startup. For instance, OpenAI must share a significant portion of its revenue with Microsoft. That's billions of dollars. One could reasonably argue that this might be why Sam Altman bangs on about OpenAI getting close to AGI soon. Many other experts in the AI field don't talk about this much, or they think the AGI debate is off base in various ways, or just not that important. Even Anthropic CEO Dario Amodei, one of the biggest AI boosters on the planet, doesn't like to talk about AGI. Microsoft CEO Satya Nadella sees things very differently. Wouldn't you? If another company is contractually required to give you oodles of money if they don't reach AGI, then you're probably not going to think we're close to AGI! Nadella has called the push toward AGI "benchmark hacking," which is so delicious. This refers to AI researchers and labs designing AI models to perform well on wonky industry benchmarks, rather than in real life. Here's OpenAI's official definition of AGI: "highly autonomous systems that outperform humans at most economically valuable work." Other experts have defined it slightly differently. But the main point is that AI machines and software must be better than humans at a wide variety of useful tasks. You can already train an AI model to be better at one or two specific things, but to get to artificial general intelligence, machines must be able to do many different things better than humans. Please help BI improve our Business, Tech, and Innovation coverage by sharing a bit about your role — it will help us tailor content that matters most to people like you. Continue By providing this information, you agree that Business Insider may use this data to improve your site experience and for targeted advertising. By continuing you agree that you accept the Terms of Service and Privacy Policy . My real-world AGI tests Over the past few months, I've devised several real-world tests to see if we've reached AGI. These are fun or annoying everyday things that should just work in a world of AGI, but they don't right now for me. I also canvassed input from readers of my Tech Memo newsletter and tapped my source network for fun suggestions. Here are my real-world tests that will prove we've reached AGI: The PR departments of OpenAI and Anthropic use their own AI technology to answer every journalist's question. Right now, these companies are hiring a ton of human journalists and other communications experts to handle a barrage of reporter questions about AI and the future. When I reach out to these companies, humans answer every time. Unacceptable! Unless this changes, we're not at AGI. This suggestion is from a hedge fund contact, and I love it: Please, please can my Microsoft Outlook email system stop burying important emails while still letting spam through? This one seems like something Microsoft and OpenAI could solve with their AI technology. I haven't seen a fix yet. In a similar vein, can someone please stop Cactus Warehouse from texting me every 2 days with offers for 20% off succulents? I only bought one cactus from you guys once! Come on, AI, this can surely be solved! My 2024 Tesla Model 3 Performance hits potholes in FSD. No wonder tires have to be replaced so often on these EVs. As a human, I can avoid potholes much better. Elon, the AGI gauntlet has been thrown down. Get on this now. Can AI models and chatbots make valuable predictions about the future, or do they mostly just regurgitate what's already known on the internet? I tested this recently, right after the US bombed Iran. ChatGPT's stock-picking ability was put to the test versus a single human analyst. Check out the results here. TL;DR: We are nowhere near AGI on this one. There's a great Google Gemini TV ad where a kid is helping his dad assemble a basketball net. The son is using an Android phone to ask Gemini for the instructions and pointing the camera at his poor father struggling with parts and tools. It's really impressive to watch as Gemini finds the instruction manual online just by "seeing" what's going on live with the product assembly. For AGI to be here, though, the AI needs to just build the damn net itself. I can sit there and read out instructions in an annoying way, while someone else toils with fiddly assembly tasks — we can all do that. Yes, I know these tests seem a bit silly — but AI benchmarks are not the real world, and they can be pretty easily gamed. That last basketball net test is particularly telling for me. Getting an AI system and software to actually assemble a basketball net — that might happen sometime soon. But, getting the same system to do a lot of other physical-world manipulation stuff better than humans, too? Very hard and probably not possible for a very long time. As OpenAI and Microsoft try to resolve their differences, the companies can tap experts to weigh in on whether the startup has reached AGI or not, per the terms of their existing contract, according to The Information. I'm happy to be an expert advisor here. Sam and Satya, let me know if you want help! For now, I'll leave the final words to a real AI expert. Konstantin Mishchenko, an AI research scientist at Meta, recently tweeted this, while citing a blog by another respected expert in the field, Sergey Levine: "While LLMs learned to mimic intelligence from internet data, they never had to actually live and acquire that intelligence directly. They lack the core algorithm for learning from experience. They need a human to do that work for them," Mishchenko wrote, referring to AI models known as large language models. "This suggests, at least to me, that the gap between LLMs and genuine intelligence might be wider than we think. Despite all the talk about AGI either being already here or coming next year, I can't shake off the feeling it's not possible until we come up with something better than a language model mimicking our own idea of how an AI should look," he concluded.
Yahoo
21-06-2025
- Business
- Yahoo
AI is changing the tech job market, and Gen Z is struggling to keep up
This post originally appeared in the BI Tech Memo newsletter. Sign up for the weekly BI Tech Memo newsletter here. As my daughter can attest, it's a tough job market out there. She's a mechanical engineer, which seems steady demand-wise. But many of her college friends are struggling to find that crucial entry-level job or internship. Some are wondering if computer science and similar courses were the right decision, with AI automating software tasks at companies. Just look at what's happening at Amazon. CEO Andy Jassy said this week that his corporate workforce will shrink over the next few years as the company uses more generative AI tools and agents. Not so coincidentally, Business Insider just spent weeks interviewing Gen Z tech job seekers as AI transforms early-career experiences across the industry. Our findings are a must-read for anyone looking to break in here and find success. Jonathan, a 26-year-old software engineer, is emblematic of what we discovered. (He's not authorized to speak to the media, so we removed his last name.) After earning his degree in 2022, he sent out nearly 300 applications and got 12 responses. It took three months to land a full-time tech role. Less than a year later, he lost that job due to his employer shutting an office. Then, he found himself repeating the process, this time applying to 600 jobs and hearing back from only five. "Everyone tells you to get into computer science. That's where all the money is at, that's where the jobs are at," Jonathan told BI. "It's a lot harder than they make it out to be." He's not alone. Recent computer science grads face a 6.1% unemployment rate in the US, higher than the national average. Wow. I never thought I'd see a stat like that. Jonathan now works in defense tech, not the video game industry he originally aspired to join. Like many of his peers, he's focused on gaining experience — any experience. It's looking like a tech career is evolving from a sure thing into something similar to every other type of job: You have to hustle and network to get your foot in the door, and then work like hell to stay employed and build steadily from there. What else should you do as an aspiring tech worker? Our Gen Z survey story doesn't have all the answers, but there's great advice in here. Read the original article on Business Insider
Business Insider
18-06-2025
- Entertainment
- Business Insider
I vibe coded a website with my daughter using AI. We're complete novices. Here's what happened.
I write a newsletter for Business Insider called Tech Memo. There's a section called AI Playground where I use a new generative AI tool every week. My latest experiment took me and my daughter deep into the world of AI coding tools. We got started on a Sunday on the couch in early June. My daughter Tessa was back home from Wake Forest for part of the summer. She has a soap business called Scrub Club and wanted a website to promote and sell her products. So we logged into and signed up for the company's Hackathon challenge. This gave us 10 million tokens to use for free. Normally, a subscription like this would cost $20 a month. Bolt is designed for novices, like us, who have no software coding experience. It's part of a new breed of AI tools that have spawned the term " vibe coding." Here's how our vibe sessions went. Spoiler alert: The Scrub Club website is up and running. Let us know what you think: Getting started We started by uploading some photos of Tessa's soaps, along with a rough description of the website she wanted. Bolt created the site within a couple of minutes, added the soap photos to the correct sections of the site, and updated us: "Your collection now showcases an amazing variety of unique shapes, colors, and scents — from elegant leaf-impressed bars with beautiful gradient effects to whimsical unicorns and vibrant autumn maple leaves. Each product maintains the handcrafted artisan quality that makes Scrub Club special!" This was an endearing feature of Bolt. No matter how crap our ideas or requests were, the tool retained a positive and supportive attitude. Sometimes, it was overly eager, in fact. Mistakes and stuff that wasn't true When we edited some of the product listings, it took the header image of the Scrub Club logo and incorrectly put it as a new product listing. We had to type in another request to remove that. And as the afternoon progressed, the more we tweaked and fiddled, the more odd updates and mistakes the tool made. The Bolt tool added some things that were not entirely true, to put it mildly: It listed bestsellers that were not the top-selling products It added customer reviews, saying things like "14 ***** Reviews" when there have been very few reviews online for Tessa's soaps so far. It also kept insisting on saying Scrub Club had "+100 Happy Customers." The real number (so far!) is just over 20. Bolt was particularly over-eager when it came to describing the soaps Tessa made and listed on the Scrub Club site. For one of her blue soap bars, it mentioned "ocean breeze scent." Tessa has not managed to bottle the smell of the sea yet for her products. Bolt also highlighted "lavender and vanilla fragrance," "candy and berry scent," and decided that a yellow moth-shaped soap smelled of "spice." None of this is true, although my daughter's soaps do smell nice! There's been some early research on trying to get machines to understand different smells, but that's clearly not embedded in the AI models supporting Bolt's coding tool yet. Getting prescriptive So, we asked Bolt to "Keep everything else the same, but remove references to scents and smells from the product listings." We had to be prescriptive like this because when we asked for other changes earlier, Bolt would make the tweaks but would also change other things and get them wrong, which was frustrating. This time, though, Bolt removed all mentions of scent and smell without messing up other things on the site. "The product descriptions now focus on the visual aspects like colors, shapes, and designs instead of any olfactory elements. Everything else remains exactly the same, including pricing, layout, and functionality," Bolt told us in its chirpy tone. By the evening on Sunday, we had a website with many of Tessa's soaps listed. It had Tessa's Scrub Club logo, pricing details, a contact form, and links to Tessa's Etsy store and her Scrub Club Instagram account. We wanted to add the ability to accept payments. Things got a little complicated at this point. So we put the project on pause and fired up Netflix with some pizza for the rest of the evening. By then, we had 9.4 million tokens left in our Bolt account. Integrating Stripe payments About a week later, we logged back onto and tried to add Stripe to Tessa's Scrub Club website. Stripe is the main way developers add the ability to accept card and online payments for their sites and apps. We clicked on an "Integrations" button on the top right of the Bolt page and Stripe appeared in a dropdown menu along with Supabase and GitHub. We chose Stripe and Bolt told us this: "To implement payments in your application, we'll need to use Stripe, which is the industry standard for secure payment processing. Before we can proceed with the integration, you'll need to: Create a Stripe account if you haven't already Once logged in, navigate to the Developers section in your Stripe Dashboard Get your Stripe secret key Once you have your Stripe secret key, let me know and I'll help you implement a secure payment system in your application." We were sent over to Stripe's website and created an account there, and tracked down our secret API key pretty easily on the developer page. We copied this long string of characters, then we went back to Bolt and pasted the key into the right place. Then we went back to Stripe and looked for what to do next. Things got pretty confusing at this point (well, for us anyway!). We ended up on a page that required us to create new products and set pricing, shipping, sales tax rates, and other complex stuff. Just asking Bolt to do it So, we gave up and just went back to Bolt and typed in "We added our secret API key from Stripe. What do we do now?" Bolt replied reassuringly: " Perfect! Now that you have your Stripe secret key set up, I'll integrate Stripe payments into your Scrub Club website. I'll add a complete payment system with a shopping cart, checkout process, and order confirmation." Then it did all the coding work and added the following features: A Shopping Cart System A Checkout Process Customer Information Form: Collects shipping details and contact info Order Summary: Shows all items and total before checkout Order Confirmation: Success page after order placement Key Features Quantity Management: Increase/decrease quantities in cart Remove Items: Delete items from cart Price Calculations: Automatic total calculations Email Integration, with customer orders sent directly to Tessa's email address. Bolt also described the ordering process clearly. When customers place an order: They fill out their information and review their cart The system sends an email to Tessa with all order details She can then contact them directly to arrange payment via Venmo, PayPal, etc. " This gives you full control over the payment process while providing a professional shopping experience," Bolt said. " The cart icon in the header shows the number of items, and customers can easily add products, review their cart, and place orders. The system is production-ready and will help streamline your order process!" we were also told. Deploying via Netlify We then clicked the blue "Deploy" button at the top right corner of the Bolt page. This connected us to Netlify, a company that hosts websites and apps created with Bolt. After about one minute, Bolt told us the site was deployed and gave us a URL link to check it out online. This all seemed too easy, so we double checked: "Does this website work fully now, or do we have to do anything else?" we typed the Bolt query box. It replied, saying the site was "fully functional!" and summarized everything we'd created, including a shopping cart, a checkout process, and a product catalog. It also shared a useful reminder about the payment process. The checkout page explains to customers that Tessa will contact them directly for payment, "which is perfect for a small business. No monthly fees or complex payment processing needed." Claiming our project and picking a domain One wrinkle at the end: Bolt said that to claim this project for ourselves, we had to go to Netlify's website and do a few more things. It provided a link for us to click. We followed that and were asked to create a Netlify account, which we did. Then we were taken straight to the page where we could claim the project. Note: If you don't do this, someone else might get hold of your claim link and grab your site for themselves and change it. Not good. Once the project belonged to us, we tried to buy a custom domain via Netlify. We chose which cost $15 for the first year. We were told to wait about 10 minutes, so we hung some laundry out to dry in the garden, then came back. By then, it was all done, including encryption certificates and other important stuff that we really didn't want to be bothered with. A review from my editor I sent the website to my editor Akin Oyedele and asked for feedback. Here's his review: What he liked: The photos were sharp and consistent. The soap shapes themselves were creative. The website overall looked clean and professionally done. My browser didn't warn that the site wasn't secure. Most of the links worked. What he didn't quite like: The logo was underwhelming compared to the visual quality of the photos When he tried to add more than one of each soap, he had to press the "add to cart" button multiple times. Usually, a counter with plus and minus signs pops up on that button on other websites. The heart buttons on some of the soaps didn't do anything He wished there were descriptions of how the soaps smell. At this point, I broke the news to Akin about Bolt's over-eager scent descriptions! A final tweak and thoughts In response to his feedback, we went back into Bolt and asked the tool to make the Scrub Club logo larger. It did that, but then cut off the top of the rest of the site. We got a little whiny at this point and sent this to Bolt: "You've cut off the top of the rest of the website now. Can you fix that please?" Bolt responded by saying, "You're absolutely right!" and went about addressing the problem. That took about two minutes, and then we asked it to deploy the site again to Netlify, which it did in about five minutes. At the end of all this, we had 8.9 million tokens left in the Bolt account. So we'd used 1.1 million. All in all, this was a relatively easy lift for two people with no software coding experience. When we were stumped by what to do next, we often just typed questions into Bolt without thinking too much about the prompts. This worked almost always. Sometimes, we had to repeat requests or get more specific and prescriptive, but that wasn't too much extra work. Total time spent on the project: About six hours. For two people with no coding background, the experience was surprisingly smooth — proof that AI tools like Bolt can empower anyone to build a real website.
Yahoo
13-06-2025
- Business
- Yahoo
What WWDC tells us about the future of Apple and the iPhone
This post originally appeared in the BI Tech Memo newsletter. Sign up for the weekly BI Tech Memo newsletter here. WWDC was a bit of a bust. Apple's Liquid Glass design overhaul was criticized on social media because it makes some iPhone notifications hard to read. A few jokers on X even shared a screenshot of YouTube's play button obstructing the "Gl" in a thumbnail for an Apple Liquid Glass promo. Need I say more? The more serious question hanging over this year's WWDC was not answered. When will Siri get the AI upgrade it desperately needs? Software chief Craig Federighi delivered the bad news: It's still not ready. That knocked roughly $75 billion off Apple's market value. The stock recovered a bit, but it's still badly lagging behind rivals this month. Google, OpenAI, and other tech companies are launching powerful new AI models and products at a breakneck pace. Apple is running out of time to prove it's a real player in this important field. Analyst Dan Ives is usually bullish, but even he's concerned. "They have a tight window to figure this out," Ives wrote, after calling this year's WWDC a "yawner." AI is complex, expensive, and takes a long time to get right. Apple was late to start building the needed foundational technology, such as data centers, training data pipelines, and homegrown AI chips. By contrast, Google began laying its AI groundwork decades ago. It bought DeepMind in 2014, and this AI lab shapes Google's models in profound ways today. When I was at Google I/O last month, one or two insiders whispered a phrase. They cautiously described an "intelligence gap" that could open up between the iPhone and other smartphones. Many Android phones already feature Google's Gemini chatbot, which is far more capable than Siri. If Apple's AI upgrade takes too long, this intelligence gap could widen so much that some iPhone users might consider switching. At I/O, these insiders only whispered this idea. That's because it will take something pretty dramatic to get people to give up their iPhones. This device has become a utility that we can't live without — even for the few days (weeks?) it might take to get used to an Android replacement. Still, if Apple doesn't get its AI house in order soon, this intelligence gap will keep growing, and things could get really siri-ous. Read the original article on Business Insider Error in retrieving data Sign in to access your portfolio Error in retrieving data Error in retrieving data Error in retrieving data Error in retrieving data
Yahoo
13-06-2025
- Business
- Yahoo
What WWDC tells us about the future of Apple and the iPhone
This post originally appeared in the BI Tech Memo newsletter. Sign up for the weekly BI Tech Memo newsletter here. WWDC was a bit of a bust. Apple's Liquid Glass design overhaul was criticized on social media because it makes some iPhone notifications hard to read. A few jokers on X even shared a screenshot of YouTube's play button obstructing the "Gl" in a thumbnail for an Apple Liquid Glass promo. Need I say more? The more serious question hanging over this year's WWDC was not answered. When will Siri get the AI upgrade it desperately needs? Software chief Craig Federighi delivered the bad news: It's still not ready. That knocked roughly $75 billion off Apple's market value. The stock recovered a bit, but it's still badly lagging behind rivals this month. Google, OpenAI, and other tech companies are launching powerful new AI models and products at a breakneck pace. Apple is running out of time to prove it's a real player in this important field. Analyst Dan Ives is usually bullish, but even he's concerned. "They have a tight window to figure this out," Ives wrote, after calling this year's WWDC a "yawner." AI is complex, expensive, and takes a long time to get right. Apple was late to start building the needed foundational technology, such as data centers, training data pipelines, and homegrown AI chips. By contrast, Google began laying its AI groundwork decades ago. It bought DeepMind in 2014, and this AI lab shapes Google's models in profound ways today. When I was at Google I/O last month, one or two insiders whispered a phrase. They cautiously described an "intelligence gap" that could open up between the iPhone and other smartphones. Many Android phones already feature Google's Gemini chatbot, which is far more capable than Siri. If Apple's AI upgrade takes too long, this intelligence gap could widen so much that some iPhone users might consider switching. At I/O, these insiders only whispered this idea. That's because it will take something pretty dramatic to get people to give up their iPhones. This device has become a utility that we can't live without — even for the few days (weeks?) it might take to get used to an Android replacement. Still, if Apple doesn't get its AI house in order soon, this intelligence gap will keep growing, and things could get really siri-ous. Read the original article on Business Insider
