\n<\/p>\n
OpenAI Announces a New AI Model, Code-Named Strawberry, That Solves Difficult Problems Step by Step<\/strong><\/a> Google Says It\u2019s Made a Quantum Computing Breakthrough That Reduces Errors<\/strong><\/a> Driverless Semis Could Be Months Away<\/strong><\/a> Geothermal Energy Could Outperform Nuclear Power<\/strong><\/a> CRISPR-Enhanced Viruses Are Being Deployed Against UTIs<\/strong><\/a> OpenAI\u2019s Fund-Raising Talks Could Value Company at $150 Billion<\/strong><\/a> Transistor-Like Qubits Hit Key Benchmark<\/strong><\/a> Inside Google\u2019s 7-Year Mission to Give AI a Robot Body<\/strong><\/a> The AI Spending Spree, in Charts<\/strong><\/a> Picture Credit score:\u00a0Laura Skinner<\/a> \/\u00a0Unsplash<\/a><\/em><\/p>\n<\/div>\n
Will Knight | Wired
<\/em>\u201cThe corporate at present introduced a brand new advance that indicators a shift in method\u2014a mannequin that may \u2018cause\u2019 logically by many troublesome issues and is considerably smarter than present AI with no main scale-up. The brand new mannequin, dubbed OpenAI o1, can resolve issues that stump present AI fashions, together with OpenAI\u2019s strongest present mannequin, GPT-4o.\u201d<\/p>\n
Sophia Chen | MIT Expertise Evaluate
<\/em>\u201cOne main problem has been that quantum computer systems can retailer or manipulate info incorrectly, stopping them from executing algorithms which are lengthy sufficient to be helpful. The brand new analysis from Google Quantum AI and its tutorial collaborators demonstrates that they’ll really add parts to scale back these errors. \u2026\u2019This error correction stuff actually works, and I feel it\u2019s solely going to get higher,\u2019 wrote Michael Newman, a member of the Google workforce, on X.\u201d<\/p>\n
Timothy B. Lee | Ars Technica
<\/em>\u201cOn a sunny morning in December, an 18-wheeler will pull right into a truck depot in Palmer, Texas, simply south of Dallas. The driving force will step out of the cab and assist switch his trailer to a second rig outfitted with highly effective sensors. This second truck will head south on Interstate 45 towards Houston. \u2026Vans journey the 200 miles between Dallas and Houston on a regular basis. However there shall be one thing particular concerning the center leg of this journey: There shall be nobody within the car.\u201d<\/p>\n
Editorial Employees | The Economist
<\/em>\u201cHow huge might EGS [or enhanced geothermal systems] get? Large enough. Although\u00a0DOE analyses counsel solely round 40GW\u00a0of typical geothermal useful resource exist in America, new strategies broaden the theoretical potential to a whopping 5,500GW\u00a0throughout a lot of the nation, with sturdy potential in over half of states. The warmth is unquestionably on.\u201d<\/p>\n
Emily Mullin | Wired
<\/em>\u201cThe worldwide rise<\/span>\u00a0in\u00a0antibiotic resistance\u00a0is making bacterial infections tougher to deal with and rising the chance of illness unfold, extreme sickness, and dying. As soon as thought-about miracle medicine,\u00a0antibiotics\u00a0at the moment are shedding their effectiveness in opposition to ever-evolving\u00a0micro organism. One firm is aiming to deal with infections with a distinct technique: arming tiny viruses known as\u00a0bacteriophages with CRISPR.\u201d<\/p>\n
Cade Metz, Mike Isaac, Tripp Mickle, and Michael J. de la Merced | The New York Instances
<\/em>\u201cIf the brand new deal is accomplished, OpenAI shall be extra invaluable than SpaceX, the personal rocket firm based by Elon Musk (who was additionally one in all OpenAI\u2019s co-founders). It might even be almost twice as invaluable as Intel, the venerable chip large, whose whole market worth has slipped to round $83 billion because it has been unable to maintain up with the AI growth.\u201d<\/p>\n
Dina Genkina | IEEE Spectrum
<\/em>\u201cA workforce in Australia has\u00a0just lately demonstrated\u00a0a key advance in metal-oxide-semiconductor-based (or MOS-based) quantum computer systems. They confirmed that their two-qubit gates\u2014logical operations that contain a couple of quantum bit, or qubit\u2014carry out with out errors 99 p.c of the time. \u2026What\u2019s extra, these MOS-based quantum computer systems are appropriate with present CMOS expertise, which can make it extra easy to fabricate numerous qubits on a single chip than with different strategies.\u201d<\/p>\n
Hans Peter Brondmo | Quote
<\/em>\u201cAs the top of Alphabet\u2019s AI-powered robotics moonshot, I got here to consider many issues. For one, robots can\u2019t come quickly sufficient. For one more, they shouldn\u2019t seem like us. \u2026I’m extra satisfied than ever that the robots want to come back. But I’ve considerations that Silicon Valley, with its give attention to \u2018minimal viable merchandise\u2019 and VCs\u2019 basic aversion to investing in {hardware}, shall be affected person sufficient to win the worldwide race to present AI a robotic physique. And far of the cash that’s\u00a0being invested is specializing in the incorrect issues. Right here is why.\u201d<\/p>\n
Nate Rattner and Tom Dotan | The Wall Road Journal
<\/em>\u201cGenerative synthetic intelligence has sparked one of many greatest spending booms in fashionable American historical past, as firms and buyers guess lots of of billions of {dollars} that the expertise will revolutionize the worldwide economic system and someday result in huge income. The query is when, and even whether or not, all these investments will repay.\u201d<\/p>\n