“All of these models have gotten quite complex,” he wrote in response to a question about why OpenAI’s next AI models were taking so long. “We also face a lot of limitations and hard decisions about [how] we allocated our compute towards many great ideas.”

Many reports suggest that OpenAI has struggled to secure enough compute infrastructure to run and train its generative models. Just this week, Reuters, citing sources, said that OpenAI has for months been working with Broadcom to create an AI chip for running models, which could arrive as soon as 2026.

Partly as a result of strained capacity, Altman said, OpenAI’s realistic-sounding conversational feature for ChatGPT, Advanced Voice Mode, won’t be getting the vision capabilities first teased in April anytime soon. At its April press event, OpenAI showed the ChatGPT app running on a smartphone and responding to visual cues, such as the clothes someone was wearing, within view of the phone’s camera.

Article

On Wednesday, Finance Minister Rachel Reeves announced a move to hike capital gains tax (CGT) — a levy on the profit investors make from the sale of an investment — as part of a far-reaching announcement on the Labour government's fiscal spending and tax plans.

The lower capital gains tax rate was increased to 18% from 10%, while the higher rate climbed to 24% from 20%. Reeves said the increases will help bring in £2.5 billion ($3.2 billion) of additional capital to the public purses.

It was also announced that the lifetime limit for business asset disposal relief (BADR) — which offers entrepreneurs a reduced rate on the level of tax paid on capital gains resulting from the sale of all or part of a company — would sit at £1 million.

She added that the rate of CGT applied to entrepreneurs using the BADR scheme will increase to 14% in 2025 and to 18% a year later. Still, Reeves said the U.K. would still have the lowest capital gains tax rate of any European G7 economy.

Article

"It's an important milestone," Apple CFO Luca Maestri said on a call with analysts. "We've got to a run rate of $100 billion. You look back just a few years ago and the the growth has been phenomenal."

Apple first broke out its services revenue in the December quarter of 2014. At the time, it was $4.8 billion.

Apple's services unit has become a critical part of Apple's appeal to investors over the past decade. Its gross margin was 74% in the September quarter compared to Apple's overall margin of 46.2%.

Services contains a wide range of different offerings. According to the company's SEC filings, it includes advertising, search licensing revenue from Google, warranties called AppleCare, cloud subscription services such as iCloud, content subscriptions such as the company's Apple TV+ service, and payments from Apple Pay and AppleCare.

Article

Jassy added that Amazon continues to “re-architect the brain” of Alexa with “a new set of foundation models” that the company plans to reveal “in the near future.”

Amazon, which first announced that it’d revamp Alexa with generative AI technologies in 2023, is said to be replacing its own Alexa-powering models with Anthropic’s after encountering technical challenges. (Amazon is a major investor in Anthropic.) At one point during the Alexa redesign, the unreleased, upgraded assistant reportedly struggled to turn on smart lights and took up to six seconds to respond to queries.

The new Alexa, code-named “Remarkable Alexa” internally, will reportedly cost $5 to $10 per month, offered alongside a less capable free plan. Some reports indicated that it would arrive in October, but it’s seemingly suffering delays.

Bloomberg reports the timeline has slipped into 2025.

Despite being in over a half billion devices worldwide, Alexa hasn’t contributed meaningfully to Amazon’s bottom line. The company has lost tens of billions of dollars in its devices business since 2017, according to The Wall Street Journal.

Article

“This is an attempt to eliminate the leading American drone company and deepen the world’s dependence on Chinese drone suppliers,” Skydio CEO Adam Bry said in a letter to customers first reported by the Financial Times.

The company, which provides drones to Ukraine’s military and Taiwan’s National Fire Agency, manufactures its products in the United States but still relies on a global supply chain for many of its components; one of the most critical components, batteries, are still sourced in China. As a result of the sanctions, which bar Chinese companies from doing business with Skydio, customers will be limited to one battery per drone, Skydio told customers.

While Skydio has a “substantial stock” of batteries on hand, it doesn’t anticipate new sources coming online until spring of next year. In the meantime, the company is extending the software license, warranty, and support term for orders affected by the battery ration by the length of time it takes for the full complement of batteries to be delivered.

Article

On a conference call with analysts following the company's third-quarter earnings report on Thursday, Jassy pointed to the success of Amazon's cloud computing business, Amazon Web Services, which has become a crucial profit engine despite the extreme costs associated with building data centers.

"I think we've proven over time that we can drive enough operating income and free cash flow to make this a very successful return on invested capital business," Jassy said. "We expect the same thing will happen here with generative AI."

Amazon spent $22.6 billion on property and equipment during the quarter, up 81% from the year before. Jassy said Amazon plans to spend $75 billion on capex in 2024 and expects an even higher number in 2025.

The jump in spending is primarily being driven by generative AI investments, Jassy said. The company is rushing to invest in data centers, networking gear and hardware to meet vast demand for the technology, which has exploded in popularity since OpenAI released its ChatGPT assistant almost two years ago.

Article

Google must obtain local content certification before resuming sales, Industry Ministry spokesperson Febri Hendri Antoni Arief told local reporters. “The local content rule and related policies are made for fairness for all investors that invest in Indonesia, and for creating added value and deepening the industry structure here,” Hendri was quoted as saying.

The ban follows Indonesia’s block on iPhone 16 sales last week after Apple failed to meet a $95 million investment commitment. Major smartphone makers must manufacture devices, develop firmware, or invest in local innovation to meet Indonesia’s content rules.

The Indonesian rule requires tech companies to source 40% of handset and tablet components domestically, a requirement that can be met through local manufacturing, firmware development or direct investment in innovation projects.

Companies can satisfy the requirements through different routes. Samsung and Xiaomi, for instance, have established manufacturing facilities, while Apple has opted to open developer academies.

Article

this country is just banning everything now 🤦

It is a new form of compute. The basic bring a completely new data structure. The vector database and how it interacts is completely new as compared to traditional databases.

Called Oasis, the model, which is available for download, powers a demo on Decart’s site: a Minecraft-like game that’s generated on the fly, end to end. Trained on videos of Minecraft gameplay, Oasis takes in keyboard and mouse movements and generates frames in real time, simulating physics, rules, and graphics.

Oasis is part of an emerging category of generative AI models called “world models.” Many of these models can simulate games — but few at frame rates as high as Oasis.

I tried the demo out of curiosity, and I’d say it has a ways to go before it’s a genuinely fun experience. The resolution is quite low, and Oasis tends to quickly “forget” the level layout — I’d turn my character around only to see a rearranged landscape.

Sparta’s primary offering is the health platform Trinsic, which tracks health vitals for enterprise clients. The technology will be integrated into the hardware firm’s B2B offering, Oura Business.

Business applications have played an increasingly important role in the world of health-based wearables in recent years. Oura Business lets employers purchase the ring in bulk or through gift cards, allowing employees to choose their specific size and color.

Article

Technology for people to thrive

We’re on a mission to equip organizations that care about their people with sustainable solutions
to maximize human readiness.

Trinsic Human Data Platform: An Expansion of the Sparta Software Corp

A natural evolution of our products and capabilities: The Trinsic Human Data Platform.

We are excited to announce the expansion of the Sparta Software Corporation ecosystem – introducing the Trinsic Human Data Platform.

Here's how the company did in comparison with LSEG consensus:

Earnings per share: 17 cents adjusted vs. loss of 2 cents expected
Revenue: $13.28 billion vs. $13.02 billion expected
Intel's revenue declined 6% year over year in the fiscal third quarter, which ended Sept. 28, according to a statement. The company registered a net loss of $16.99 billion, or $3.88 per share, compared with net earnings of $310 million, or 7 cents per share, in the same quarter a year ago.

As part of a cost reduction plan, Intel recognized $2.8 billion in restructuring charges during the quarter. There was also $15.9 billion in impairment charges tied in part to accelerated depreciation for Intel 7 process node manufacturing assets and goodwill impairment in the Mobileye unit.

The company is carrying out one of the most seminal restructuring processes since its establishment in 1968, CEO Pat Gelsinger said on a conference call with analysts.

Article

No

Here's how the iPhone maker did versus LSEG consensus estimates for the quarter ending Sept. 28:

Earnings per share: $1.64, adjusted, versus $1.60 estimated
Revenue: $94.93 billion vs. $94.58 billion estimated
iPhone revenue: $46.22 billion vs. $45.47 billion estimated
Mac revenue: $7.74 billion vs. $7.82 billion estimated
iPad revenue: $6.95 billion vs. $7.09 billion estimated
Other Products revenue: $9.04 billion vs. $9.21 billion estimated
Services revenue: $24.97 billion vs. $25.28 billion estimated
Gross margin: 46.2% vs. 46.0% estimated

Overall iPhone revenue grew 6%, in the first sign of how the iPhone 16 is faring in the market. Apple's newest devices came out Sept. 20, giving Apple about a week of new product sales in the quarter. It's still Apple's most important product, accounting for nearly 49% of the company's overall sales.

Sales of the iPhone 15 were "stronger than 14 in the year-ago quarter, and 16 was stronger than 15," Apple CEO Tim Cook told CNBC's Steve Kovach.

Article

wait so🤔 they made profit but went down in market?

The pause comes after the company reported disappointing third-quarter sales. The auto giant sold 23,509 electric vehicles during the third quarter versus General Motors’ 32,095.

Ford remains close behind GM in overall electric vehicle sales this year, reporting 67,689 sold compared with GM’s 70,450.

Both companies’ EV sales stand to fall in the coming months because of increased competition. Earlier in October, Cox Automotive reported that Tesla’s Cybertruck was the third best-selling EV on the market, behind two other Tesla vehicles.

Article

an unfortunate news but it can I say it's because of the booming automobile industry in China for EV

Not if they dont have the keys. Quantum computers might be able to hack them someday.

That said, some feel Bitcoin was developed by the NSA. If that is the case, then it already can access the wallets.

Huawei recently bucked the trend with the release of Harmony OS, though that was a direct product of geopolitical restrictions on using American products. Despite its massive resources, however, the Chinese electronics giant struggled tremendously as it scrambled to build its own Android alternative.

Given its penchant for shaking up the market, perhaps it’s not wholly surprising that London-based Nothing could build its own mobile OS from the ground up. At TechCrunch Disrupt 2024 on Wednesday, founder and CEO Carl Pei confirmed that the company is exploring what a true Nothing OS could look like.

Pei praised smartphones as “our most important gateway to the people we care about and the information we need to consume,” while criticizing the industry’s Google/Apple duopoly.

“We’re thinking about how we maneuver here,” he added, “and maybe create something of our own. Some kind of operating system.”

Article

The artificial intelligence portion of AWS is in the billions of dollars in annualized revenue, more than doubling year over year, Amazon CEO Andy Jassy, who previously led AWS, said on a call with analysts.

"I believe we have more demand than we could fulfill if we had even more capacity today," Jassy said. "I think pretty much everyone today has less capacity than they have demand for, and it's really primarily chips that are the area where companies could use more supply."

Article

How do Amazon Web Services' profit margins compare with those of its main competitors like Microsoft Azure and Google Cloud?

Amazon's online advertising business is still a fraction of the company's overall business, but its growth over the years has made it a major competitor to Alphabet and Meta, which lead the digital advertising market. Alphabet's Google currently represents 27.7% of the worldwide digital advertising market, followed by Meta at 22.8% and Amazon with 8.8%, according to data provided to CNBC by Emarketer.

Meta's third-quarter advertising revenue came in at $39.9 billion, which was up 19% compared with the year prior. That was slightly ahead of analysts' expectations of $39.49 billion, according to StreetAccount. Ads accounted for 98.3% of Meta's overall third-quarter revenue.

Alphabet generated $65.85 billion in third-quarter ad revenue, the company reported Tuesday. That was up 10% from $59.65 billion the year prior. Additionally, advertising sales for the company's YouTube unit rose 12% year over year to $8.92 billion.

Article

How does Amazon's advertising revenue compare to that of its competitors, such as Google and Facebook?

7

This event sums up exactly why Elon is such a polarizing character. Because on one hand, this is the visionary who dares to think it's possibly feasible to shoot off rockets and land them back on barge ships in the ocean. He's going to do it all again and completely revolutionize road transport. What we've learned over the last decade or so is that it may be more difficult to put a safe self-driving car on the road than to land rocket ships back here on Earth after shooting them off into space. It just speaks to the sort of level of technical challenge and capability that is going to be required for these autonomous vehicles to be acceptable to society. And it reminds me of something somewhat related - watching him on stage there, after having just seen him on stage with Trump in Butler, Pennsylvania, I will say this: he was much more in his element here.

FROM THE HEART OF WHERE INNOVATION, MONEY, AND POWER COLLIDE, FROM SILICON VALLEY AND BEYOND, THIS IS BLOOMBERG TECHNOLOGY WITH CAROLINE HYDE AND ED LUDLOW.

ED: Live from San Francisco to TV and radio audiences around the world, welcome to a special edition of Bloomberg Technology. I’m Ed Ludlow. In a few moments, CEO Pat Gelsinger will join us for a live interview following the latest earnings report. The company is giving a fourth-quarter revenue forecast slightly above estimates, which is spiking optimism that Intel is capable of reclaiming lost market share in key categories. The shares are really surging.

It’s important to note that two other giants of the technology world also reported numbers this evening. Apple, the world’s most valuable company, is down two percentage points. It grew in every geography around the world apart from Greater China. iPhone revenue overall reached a strong forecast for the final four months of the year. There’s a solid indication of Amazon regaining momentum with AWS returning to growth.

We will check in on all three names in the next 30 minutes, but now the focus is on Intel. It had a painful third quarter, and it’s giving us a signal about what might come next. First, we will go to Bloomberg Senior Executive Editor Tom Giles, who leads technology coverage in the newsroom.

TOM: Where to start? Intel has moved on from a painful third quarter and is trying to tell us it has regained footing going forward.

ED: Slightly is the key here. A lot of what they did in the most recent quarter involved cutting costs, expenses, and headcount. They took a step back on investor payouts. Much of it is about retrenching. They did give a forecast for the fourth quarter that slightly exceeded analyst expectations, which was enough to get the stock moving. It is definitely up in after-hours trading. I look at the report and see they are not out of the woods yet by any stretch of the imagination. Their orders for AI chips still lack significant momentum, and they won’t meet the $500 million threshold that they had talked about.

We are still wondering whether there will be interest from outside companies after the election. Bloomberg has reported on Qualcomm’s potential interest in an Intel deal, but it seems they will put that off until after the election, so we may not see much movement until after the new year. Investors are breathing a sigh of relief, but Gelsinger still has a lot of expectations to live up to.

ED: I want to bring in Bloomberg's Ian King, who has led semiconductor coverage in the Bloomberg newsroom since 1998. For that whole time, Intel has led in certain products, particularly chips that go into PCs and data centers. Their third quarter was painful, with billions of dollars in impairment charges and headcount reductions. Can you explain why it was so important for the company to take those actions?

IAN: Pat Gelsinger spelled it out. He said, “We are resizing the company for the size of revenue we expect going forward.” He indicated they expect growth of 3% to 5%, and if they push it, maybe they will reach the 7% range. This is a long way off from Intel’s previous performance, particularly in the AI chip market.

ED: You spoke to Pat Gelsinger on the phone briefly. Their business is split into two: chips for PCs and data centers, and manufacturing. Gelsinger explained the plan to run them as distinct divisions but still under one company.

IAN: He talked about the benefits of splitting them up while still keeping them together. He was asked about this on the call, and people still struggle with it. If you are splitting them up, why not go all the way and split the company? He insisted they need to manage their overall finances, and the majority of the work done in those factories will be their chips for the next couple of years.

ED: Those are the numbers Intel posted. Let’s get a reaction from the investment community. JoAnne Feeney is a partner and portfolio manager at Advisors Capital Management, currently with no exposure to Intel, but someone who has covered semiconductors for many years at multiple firms. What is your reaction to the Intel print and what they told us about their progress?

JOANNE: They made a lot of progress. It’s an important step forward. They had a lot of cost-cutting to do, and it seems like they got a lot done last quarter. We are seeing a big charge on the diluted EPS. However, they have a long way to go because they are taking a huge hit on their gross margin by outsourcing. That reflects a lack of execution in the past on the manufacturing side. It is not just coming to an end. We did not get good news about that this quarter. In fact, it is really not going to play through until 2026 when they ramp Panther Lake in real volume.

Even Pat Gelsinger said on the call that 70% will be in-house, but 30% will not. I think that is a smart business decision; they have to be a reliable provider of the chips. If there’s any hiccup in the yields, they will get on Panther Lake like they had better have. They have gone from 40% to now underperforming. That is because they have not executed. I have always said if someone can right the ship, it is Pat Gelsinger, a former Intel guy back at the helm. He and Dave Kinzer have made incredible progress this quarter.

JOANNE: They are a distant follower. It will take them a long time to make inroads. It’s not just NVIDIA and AMD with new products; it’s custom silicon coming out of the cloud providers themselves, like Google and OpenAI working with Broadcom. This reflects the different opportunities Intel has. It’s not just AI accelerators; it’s that they have lost so much market share in servers to AMD. They used to be at 97%.

ED: JoAnne Feeney, stay with us. I want to go back to Ian King. Historically, Intel was the leader in product categories and margins. It was a company with margins of 60% or above, and that is not the case anymore. Why?

IAN: Historically, their massive advantage was having the best factories in the industry, which cost tens of billions of dollars to build. If they are good and give you better products, that’s an asset. Right now, it’s the reverse; they are not the best, and that’s laying on the margins.

ED: What were the answers?

IAN: Two years.

ED: What do you mean?

IAN: There’s this foundry business, and part of it is serving Intel, their own biggest customer. They’ve named impressive people to lead the manufacturing of chips for other companies. Now, does there seem to be a timeline building of when that might happen?

IAN: The optimism is that now we have made our factories great again. Now they are good, and it will help our products attract outside money. The analysts are saying, “When?” What’s happening there? As JoAnne said, they are outsourcing a big chunk of their best stuff, which has tremendous margins, and there’s a negative impact if you have factories and you aren’t making your best stuff. That’s not how you want to be in the industry.

ED: When I moved to California six years ago, Intel was part of the lore, the history of Silicon Valley. They made very painful decisions. When I posted on social media that Pat Gelsinger was doing this special program, many voiced their support for him. What would your question be for him?

IAN: How will you bring that gross margin up? How will you convince potential foundry customers they will get priority in wafer starts when they need them? That has always been the challenge in years past when Intel has talked about starting a foundry business. The concern among potential customers was, “How do we know you will get the wafer starts you promised? How do we know you won’t prioritize your own products?” Secondly, do they have the extensive libraries they need to build chips for other companies? That is a set of libraries that have obviously built up over many, many years and decades at TSMC and Samsung, but it’s new for Intel. Can they execute to build other companies' products as well as they do for their own?

ED: JoAnne Feeney of Advisors Capital Management and Bloomberg’s Ian King, thank you. We are joined by Pat Gelsinger live on Bloomberg Television and radio. Pat, thank you for your time. We were just talking about how you made painful decisions that showed up as impairment charges and reflected on the bottom line. Can I ask if that’s it now? Are the actions taken, and do you have a clear line of sight on what you want to do, or will there be more restructuring to come?

PAT: We worked very hard this quarter to get this done. The people actions and the restructuring charges were largely finished this quarter. It was a challenging quarter that way. But to deliver better-than-expected results—if we eliminate one-time charges and take guidance up for Q4—I’m proud of our team for being able to do

A new development in AI “swarms” serves as a wake up call for everyone involved in cybersecurity, automation and, in fact, IT generally: OpenAI’s Swarm.

What is OpenAI Swarm?
OpenAI launched an experimental framework last month called Swarm. It’s a “lightweight” system for the development of agentic AI swarms, which are networks of autonomous AI agents able to work together to handle complex tasks without human intervention, according to OpenAI.

Swarm is not a product. It’s an experimental tool for coordinating or orchestrating networks of AI agents. The framework is open-source under the MIT license (which allows Python developers to use, modify, and distribute the software with minimal restrictions), and available on GitHub.

In the GitHub readme section, OpenAI says:

“Swarm is currently an experimental sample framework intended to explore ergonomic interfaces for multi-agent systems. It is not intended to be used in production, and therefore has no official support. (This also means we will not be reviewing PRs or issues!)

The primary goal of Swarm is to showcase the handoff & routines patterns explored in the Orchestrating Agents: Handoffs & Routines cookbook. It is not meant as a standalone library and is primarily for educational purposes.”

Swarm is not totally unique. Other existing systems can be used for the orchestration of multiple agents, which approaches the functioning of agentic AI swarms. Though not explicitly designed for swarming, they can be used for making AI agents interact with each other to varying degrees. These include: Microsoft AutoGen, CrewAI, LangChain, LangGraph, MetaGPT, AutoGPT, and Haystack.

While Swarm might be designed for simplicity and relative ease of use, all these other tools are more robust, reliable, supported and ready for prime-time.

OpenAI apparently launched Swarm to explore methods for improving agent collaboration through “routines” and “handoffs.” In this case, “routines” are predefined sets of instructions that guide agents through tasks or workflows. They serve as recipes for agents to follow, which adds control and predictability to multi-agent systems. “Handoffs” enable one agent to delegate a job to another based on the current context. For example, if the agent requires something specific that can be better handled by an agent specializing in that task, it can delegate it. That “handoff” provides the history of the task to the new agent, so it has context under which to proceed.

Article

Soon the tech companies may be empty with a lot of AI servers and few stuff monitoring the automation. We'll advance faster with this kind of tech

Just to explain for someone who is newbie. How would you explain what a agentic AI, and how does it differ from traditional AI systems?

However, a new peer-reviewed study led by Sydney-based researchers Stephen Woodcock and Jay Falletta has found that the time it would take for a typing monkey to replicate Shakespeare's plays, sonnets and poems would be longer than the lifespan of our universe.

Which means that while mathematically true, the theorem is "misleading", they say.

As well as looking at the abilities of a single monkey, the study also did a series of calculations based on the current global population of chimpanzees, which is roughly 200,000.

The results indicated that even if every chimp in the world was enlisted and able to type at a pace of one key per second until the end of the universe, they wouldn't even come close to typing out the Bard's works.

There would be a 5% chance that a single chimp would successfully type the word "bananas" in its own lifetime. And the probability of one chimp constructing a random sentence - such as "I chimp, therefore I am" - comes in at one in 10 million billion billion, the research indicates.

“It is not plausible that, even with improved typing speeds or an increase in chimpanzee populations, monkey labour will ever be a viable tool for developing non-trivial written works,” the study says.

The calculations used in the paper are based on the most widely accepted hypothesis for the end of the universe, which is the heat death theory.

Despite its name, the so-called heat death would actually be slow and cold.

In short, it's a scenario in which the universe continues to both expand and cool - while everything within it dies off, decays, and fades away.

“This finding places the theorem among other probability puzzles and paradoxes... where using the idea of infinite resources gives results that don’t match up with what we get when we consider the constraints of our universe,” Associate Prof Woodcock said in a statement about the work.

News headline

Finding it here is one of the first detections of hydrogen sulphide on an exoplanet - or planet outside of our solar system.

"So, if your nose could work at 1000C ... the atmosphere would smell like rotten eggs," said Dr Guangwei Fu, an astrophysicist at Johns Hopkins who led the research.

The study was published in the journal Nature.

Although hydrogen sulphide is one of the gasses that indicate that distant planets might be home to alien organisms, researchers aren't searching for life on this planet because it's a gas giant, like Jupiter, and too hot.

However, they say finding hydrogen sulphide here is a step toward understanding how planets form.

James Webb is opening up a new vista in analysing the chemicals in faraway planets and so helping astronomers learn more about their origins.

"It's been quite a game changer. It's really revolutionising the field of astronomy. It has delivered the capability promised and it even exceeded our expectations in certain ways," said Dr Fu.

The researcher plans to use information from the space telescope to study more planets.

News headline

Webb used its huge 6.5m-wide primary mirror and sensitive infrared instruments to make the discovery.

The telescope's previous record holder was a galaxy seen at 325 million years after the Big Bang.

Astronomers say the most interesting aspect of the latest observation is not so much the great distance involved - as amazing as that is - but rather the size and brightness of JADES-GS-z14-0.

Webb measures the galaxy to be more than 1,600 light years across. Many of the most luminous galaxies generate the majority of their light via gas falling into a supermassive black hole. But the scale of JADES-GS-z14-0 indicates that is not the explanation in this case. Instead, the researchers believe the light is being produced by young stars.

"This much starlight implies that the galaxy is several hundreds of millions of times the mass of the Sun! This raises the question: how can nature make such a bright, massive, and large galaxy in less than 300 million years?" said Webb astronomers Stefano Carniani and Kevin Hainline.

Dr Carniani is affiliated to Scuola Normale Superiore in Pisa, Italy, and Dr Hainline is from the University of Arizona in Tucson, Arizona.

The $10bn James Webb Space Telescope (JWST) was launched in 2021 as a joint endeavour of the US, European and Canadian space agencies.

It was designed specifically to see farther across the cosmos - and further back in time - than any previous astronomical tool.

One of its key objectives was to find the very first stars to ignite in the nascent Universe.

These giant objects, perhaps many hundreds of times the mass of our Sun, were made only of hydrogen and helium.

In JADES-GS-z14-0, Webb can see a significant amount of oxygen, which tells researchers the galaxy is already quite mature.

"The presence of oxygen so early in the life of this galaxy is a surprise and suggests that multiple generations of very massive stars had already lived their lives before we observed the galaxy," added Drs Carniani and Hainline.

The "JADES" in the object's name stands for "JWST Advanced Deep Extragalactic Survey". It is one of a number of observation programmes using the telescope to probe the first few hundred million years of the cosmos.

"z14" refers to "Redshift 14". Redshift is the term astronomers use to describe distances.

It's essentially a measure of how the light coming from a far-off galaxy has been stretched to longer wavelengths by the expansion of the Universe.

The greater the distance, the greater the stretching. The light from the earliest galaxies is stretched from ultraviolet and visible wavelengths into the infrared - the part of the electromagnetic spectrum to which James Webb's mirrors and instruments were specifically tuned.

“We could have detected this galaxy even if it were 10 times fainter, which means that we could see other examples yet earlier in the Universe - probably into the first 200 million years,” said Prof Brant Robertson from the University of California at Santa Cruz.

The JADES discovery and its implications are described in a number of scholarly papers published on the arXiv preprint service.

News headline

ChatGPT Search: Disrupting the Search Engine Market

Introduction

OpenAI has entered the search engine arena with its new ChatGPT Search feature, positioning itself as a potential game-changer in how users find and consume online information. This development signals a significant shift in the search technology landscape, challenging established players like Google and emerging AI-powered search platforms like Perplexity.

Key Features and Capabilities

Search Functionality

ChatGPT Search introduces several innovative features:

• Web Search Integration: Accessible via a globe icon in the ChatGPT interface
• Availability: Currently available for pro and team users, with plans to expand to enterprise and free users
• Unique Search Types: Built-in support for specialized searches including:
  • Weather
  • Stocks
  • Sports news
  • Maps

Technical Architecture

The search feature is powered by a fine-tuned version of GPT-4, utilizing:

• Novel synthetic data generation techniques
• Outputs from OpenAI's 0.1 preview model
• Third-party search provider integrations

Partnerships and Content Sources

OpenAI has established partnerships with an impressive array of content providers, including:

• Associated Press
• Axel Springer
• Condé Nast
• Financial Times
• Getty
• Hearst
• News Corp
• The Atlantic
• Time
• Vox Media

These partnerships ensure access to real-time, high-quality information across various domains.

Comparative Analysis: ChatGPT Search vs. Perplexity

Strengths and Differences

Information Presentation

• Perplexity: Provides more comprehensive, context-rich answers
• ChatGPT Search: Offers more concise, direct responses

Search Speed

Both platforms demonstrate comparable search speeds, with slight variations depending on the query.

Accuracy and Hallucination

• ChatGPT Search shows potential for occasional inaccuracies
• Perplexity tends to provide more detailed context, potentially reducing misinformation risks

Specific Use Case Comparisons

Example 1: Podcast Episode Search

• ChatGPT Search: Direct link to content
• Perplexity: Comprehensive background information with embedded link

Example 2: Specialized Topic Search

• ChatGPT Search: Sometimes provides quick answers without web search
• Perplexity: Consistently leverages web search for responses

Implications for the Search Market

Platform Risk for Competitors

The launch highlights significant platform risk for companies building on top of OpenAI's infrastructure. As demonstrated, OpenAI can quickly develop and integrate features that compete with existing third-party services.

Potential Impact on Google and Other Search Engines

ChatGPT Search represents a substantial challenge to traditional search engines, offering:

• AI-powered contextual understanding
• Comprehensive information synthesis
• Multi-format search capabilities

Conclusion

ChatGPT Search is more than just another search tool—it's a potential paradigm shift in information retrieval. By combining AI intelligence, extensive partnerships, and user-friendly design, OpenAI is positioning itself as a serious contender in the search engine market.

Future Outlook

As the feature rolls out to more users and continues to be refined, it will be crucial to monitor:

• Accuracy improvements
• Expansion of search capabilities
• User adoption rates
• Competitive responses from Google and other search providers

Disclaimer: This analysis is based on initial observations and may evolve as the technology develops.

"Mars One Second: Apple's Own Version of AI Now Available for iPhone"

Hey there! Rich DeMuro here to show you what Apple’s new AI can do. That new glow around Siri – that’s what it looks like when you activate Apple Intelligence. See that glow around the edge? It’s only available on certain iPhones. You need an iPhone 15 Pro or above, like the iPhone 15 Pro, 15 Pro Max, or iPhone 16 models. So, it’s limited.

Samsung introduced smartphone AI features earlier this year, and now it’s Apple’s turn. The company’s new Apple Intelligence features are rolling out, bringing a fresh take on AI for iPhone users. Let’s take a closer look!

To learn more, I went to the Apple Store, where they’re offering pop-up demos as part of their “Today at Apple” sessions. My guide, Meyer Austin, says the best part about Apple Intelligence is how intuitive and easy it is to use. To get these new features, you need an iPhone 15 Pro or an iPhone 16 model and the iOS 18.1 software update. Once you have that, go into Settings and toggle on the switch for Apple Intelligence.

Some features aren’t available just yet, but there are new writing tools that let you type what you want, and it immediately writes it for you. I love how you can hop between apps seamlessly – no third-party apps needed! These features let you highlight text and rewrite it in different styles. It works anywhere there’s a blinking cursor.

Emails have a new feature that can help you reply automatically, suggesting responses to any questions in an email, prioritizing efficiency. Apple Intelligence is all about making things easier! The Photos app has a new cleanup tool that lets you remove unwanted people or objects from your photos.

Siri is also getting smarter – you can set a timer, and the voice assistant will better understand your requests, even if you stumble. There’s also a new way to type to Siri: a double-tap at the bottom of your screen brings up a colorful glow with whimsical animations.

Apple says it’s doing all of this in a way that protects your privacy. It’s designed with privacy in mind, from the moment you open your device to any app you use – those privacy safeguards are built right in.

The full Apple Intelligence feature set will arrive in December. That’s when you’ll be able to create custom emojis, generate images, and even use Siri to interact with ChatGPT.

Before we go, let’s quickly check out the new cleanup feature. This tool, now available in Photos, lets you highlight something you want to remove – and just like that, it’s gone! It’s so cool to see it in action.

"AI: Transforming the Future of Healthcare"

There's no question that AI will play an instrumental role in shaping the future of healthcare. When we consider the impacts of AI on health, it really comes down to addressing the core challenge in healthcare today: a mismatched market. In recent years, and as we move forward, various factors are widening the gap between the quality of health people want and what the current system can actually deliver.

At its core, AI serves as a fundamental enabler—a powerful new technology and information layer capable of supporting improvements across the entire spectrum of activities within the global health system. When it comes to specific areas, there are countless use cases where AI is set to make a positive impact.

We've already seen early successes in fields like imaging, workflow augmentation, triage, and drug discovery. These areas allow us to demonstrate clear value and secure a return on investment (ROI) for organizations. This initial ROI can help fuel further progress toward longer-term AI advancements in healthcare, such as personalized medicine, robotic surgery, and end-to-end disease management—those bigger-picture goals that we aim to achieve.

"Advancing Healthcare with AI: Insights from Dr. CIP Wadhwa"

My name is Dr. CIP Wadhwa. I’m an internist and geriatrician, and I also completed training in healthcare economics at the Wharton School. I have a strong interest in population health and the use of technology and innovative payment methods to advance healthcare. I am particularly focused on reducing morbidity and preventable mortality.

I’ve had the privilege of working in both academia and regulatory roles, and recently, I’ve become more involved in global health and national models of care. I’m especially excited to collaborate in the Gulf region to advance population health.

One area where conversational AI, like front-end speech recognition combined with ambient clinical documentation, shows promise is in creating new efficiencies for clinicians. This technology helps reduce the time clinicians spend documenting in the electronic health record, allowing them to focus more on patient care rather than on data entry. With ambient clinical documentation, we can record conversations between the patient, physician, and family, and then use generative AI to create an initial draft of the clinical note. The note becomes a byproduct of the encounter, physical exam, and conversation, enabling clinicians to focus on the patient, make eye contact, and pick up on non-verbal cues instead of typing and clicking.

I believe this approach will have a measurable impact on patient experience, as it allows clinicians to concentrate on the patient and the family. This technology captures essential clinical details from conversations needed to document an accurate, comprehensive picture of the patient's health, and enables the physician to take on a review-and-edit role, enhancing productivity, quality of care, and the patient experience.

However, new technology should be approached with caution. While some early adopters may quickly become comfortable with it, a phased rollout with champions within the organization can help address concerns. While technology has promised to simplify healthcare, there have been unintended consequences. Initial experiences with this technology, however, have helped alleviate some worries, and as more people start to use it, we will understand its value more clearly.

AI tools could enhance healthcare in several ways:

1. Supporting Differential Diagnosis: Although the brain is remarkable, AI-assisted differential diagnosis can offer thoughtful, point-of-care suggestions.

1. Enabling Earlier Diagnoses: By expanding the differential diagnosis earlier, especially with subtle early symptoms, AI could help bring certain conditions to mind sooner.

1. Incorporating Social and Genetic Factors: A holistic view of the patient, including social risks, genetics, and life context, could improve diagnosis and care in a structured, unobtrusive manner.

1. Supporting Public Health: While physicians primarily work on an individual level, these tools can help with the early identification of public health threats, be they infectious or environmental, allowing physicians to wear their public health hats at both micro and macro levels.

Respecting patient privacy and autonomy is essential. Patients should have the right to opt out of AI documentation if they’re uncomfortable with it. It’s also important for patients to be informed if genetic information or biomarkers are used to guide decisions. The days of simply telling patients what to do are over—engaging them in their care enhances both understanding and outcomes.

In the past two years, I’ve spent significant time in the Gulf region, particularly in the UAE, Saudi Arabia, Qatar, and Kuwait, where I’ve had the privilege of working with regulators and health systems. I am impressed by the commitment to population health, thoughtful use of technology, and inclusion of youth in the workforce. The Gulf region is at the forefront of these advancements, with both health systems and regulators adopting a technology-informed, population health approach. It’s an exciting time to be involved in healthcare in the GCC.

One of the challenges with AI is ensuring that its training datasets are free from bias. It’s critical to have explainability and transparency in AI models for both patient and physician trust. Continuous monitoring is essential to avoid unexpected shifts in model performance, commonly known as "model drift." It’s not a “set it and forget it” approach but a continuous process to ensure models remain effective and reliable. Without careful integration, these efforts may face significant challenges.

I’m assisting an adventure capital group called Pul 63, which focuses on the Philippine market. It was started by an engineer based in the Philippines who exclusively invests in healthcare startups. I asked to see their startups, and he had about 30 companies under his umbrella. One that caught my eye was a claims system, which is a very simple solution. It learned the rules of the Philippine Health Insurance Corporation, but he didn’t know how to sell it to them. So, I became their business development advisor.

I said, “Look, I think I can help you generate revenue from your product.” I structured it so that PhilHealth uses the platform as their claims management system. All claims from every provider in the Philippines are checked at the provider's site before they are sent out electronically or physically. The claims are validated by AI using PDF files, images, and data related to the procedures. The billings are all scanned PDFs that are easily fed into our cloud-based system.

Our AI Swift Claims ensures that every line item in a claim is met according to PhilHealth standards. We verify whether the claim is covered and to what percentage. For instance, do we cover 30% of a heart surgery or 20% of dialysis? We also conduct fraud checks to confirm that the person exists and that they were in the emergency ward for the reported issue.

This system is still growing, and we are expanding our fraud detection technology. We're implementing a secure ID card system outside of the cloud to prevent hacking. This way, we can store the resident's medical data on an ID card that they carry, ensuring that their information is secure.

The claims system is making a significant impact by reducing 99% of all return-to-hospital claims. As you know, hospital claims often get returned if there’s one mistake, causing significant delays in payments. In the Philippines, hospitals could wait anywhere from 90 days to two years to receive payments, leading to backlogs. Our system reduces that 90-day cycle to just 15 days. I asked PhilHealth why it couldn't be done within a day, and they explained that the banking system takes a week to a week and a half. We are currently working on automating the banking approval process to expedite fund transfers into provider accounts.

This is a huge project for us, and I refer to it as Project 28.8. The goal is to create a sustainable primary care solution to ensure that our 28.8 million public school children receive their annual check-ups and have access to primary care 24/7 in their home villages. In many rural areas, schools serve as community centers—places for voting, playgrounds, and gatherings—so we believe we can establish what we call Life Clinics or LabEx Digital Clinics in these schools.

The Life Clinic is essentially a clinic in a box, equipped with complete diagnostics. For example, we can conduct a complete CBC using AI imaging. One drop of blood goes into a reader that takes a picture, which is then sent to the cloud. We can provide a comprehensive 20-parameter CBC report, including platelet counts. I found a manufacturer in Brazil that developed an AI-enabled digital microscope to accurately read platelet counts.

The project is in its infancy; we're currently in the 100,000 range, but with approval from the Department of Education, we hope to scale up to a million children every quarter. We need the Secretary's support to ensure that all schools in the Philippines sign up.

AI is not a disruptor; it represents good medicine. I believe healthcare is one sector where AI will enable improvements—hundredfold, even thousandfold. From genetics to new medicines, AI will enhance how we monitor patients, and ultimately how we feed ourselves.

AI should be viewed as a tool for improvement, much like electricity transformed our lives without disrupting older technologies like candles. AI will enhance our education, communication, work, and healthcare. In the coming years, we will likely sequence many human genomes and match them with clinical data, revolutionizing healthcare and potentially curing diseases like cancer.

Currently, the cost of genetic sequencing is around $300 per person. I’m involved in liquid biopsy initiatives that focus on early cancer detection, using AI to assess risk levels. The future of AI, genetics, and clinical data is promising, and I firmly believe that, like electricity, AI will ultimately save more lives than it risks.