ET Intelligence Group: Shares of PNB Housing Finance have gained 24% in 2026 so far, the highest among peers, most of whom have failed to generate returns. The broader BSE Financial Services index has lost 4% year-to-date. The company’s standout performance on the bourses can be attributed to sustained momentum in loan disbursement and the lender’s growing focus on high-yield segments of affordable and emerging housing amid stable asset quality. The company is expected to report double digit annual growth in disbursements and net profit between FY26 and FY28.

Despite the recent price gain, the stock’s valuation at a trailing price-book (P/B) multiple of 1.6 remains below the two-to-three range for peers. It reflects lower return ratios due to the asset mix tilting more towards prime housing, which has lower yields compared with affordable housing segment. For PNB Housing, return on equity ranges between 11% and 13%. Some of the peers with higher P/B multiples including Aadhar Housing Finance, Aptus Value Housing Finance India and Home First Finance Company India have RoEs of 15-20%. These lenders predominantly focus on low-cost housing.

To address the valuation gap, PNB Housing has chalked out plans to increase share of the affordable and emerging housing segment in retail loan portfolio to 45% by the end of FY27 and to 50% in the next two years from over 40% at present. It also launched financing for developers and micro housing during the June quarter to improve the yield, which remained at around 9.5%, similar to the previous quarter.

Read more: Shifting Gears: PSBs borrow more to cater to credit demand as deposits lag

The lender changed the disbursement recognition method in the June quarter to cheque realisation from cheque handover basis. This resulted in a sharp sequential fall of 37% in disbursements at ₹5,882 crore though it increased by 18% year-on-year. Assets under management (AUM) and total loan book rose by 13% and 15% to ₹93,021 crore and ₹89,670 crore respectively. The gross nonperforming assets (GNPA) ratio remained under 1%, reflecting stable asset quality.

“The re-entry into developer finance, increasing mix of affordable and emerging segments and expansion into micro housing should support yields, while strong disbursement momentum and sustained recoveries underpin growth and profitability,” mentioned JM Financial Institutional Securities in a review report.

First Eagle is an independent investment management firm that manages approximately $149* billion in assets (as of 09/30/24) on behalf of institutional and individual clients. With the core purpose of providing prudent stewardship of client assets, the firm focuses on active, fundamental and benchmark-agnostic investing, with a strong focus on downside mitigation. First Eagle’s investment capabilities include equity, fixed income and multi-asset strategies. With a heritage dating back to 1864, First Eagle has helped its clients avoid permanent impairment of capital and earn attractive returns through widely varied economic cycles—a tradition that is central to its mission today. First Eagle Investments is the brand name for First Eagle Investment Management, LLC and its subsidiary investment advisers. Note: This account is not managed or monitored by First Eagle, and any messages sent via Seeking Alpha will not receive a response. For inquiries or communication, please use First Eagle’s official channels.

Chipmaker Nvidia has reported another huge jump in sales as the global push to build artificial intelligence (AI) systems continues at a rapid pace.

The company said on Wednesday it brought in $96bn (£71bn) in revenue during the second quarter, more than double from a year ago. And it expects revenue of $108bn next quarter.

“AI has reached its inflection point,” CEO Jensen Huang said in prepared remarks, describing the infrastructure buildout as going “at full steam.”

The revenue figures beat Wall Street’s expectations, leading Nvidia shares to rise about 4% in after hours trading.

The company’s data centre division alone generated $89bn last quarter, up 117% from a year ago, underscoring just how much of the industry now depends on Nvidia’s hardware.

Essentially every notable tech company building AI tools and infrastructure, including Amazon, Meta, Google, Microsoft, use Nvidia chips to do so.

Financial analysts said the strong results highlight Nvidia’s ongoing momentum.

Matt Britzman, senior equity analyst at Hargreaves Lansdown, called it “another monster set of results,” noting that revenue and earnings both topped forecasts.

He said the guidance for next quarter “points to revenue comfortably above $110bn.”

Nvidia’s growing financial strength has also reshaped its role in the sector.

It has become a backer to those that rely on its chips, providing some funding to the likes of OpenAI, Anthropic, and SpaceX to help continue the costly buildout of AI infrastructure.

Its financial success and processors are now central the AI boom, powering the data centres used to train and run AI models.

Demand for that computing muscle has helped transform Nvidia into the world’s most valuable firm, with a market capitalisation above $5tn.

Competition is emerging – from customers designing their own processors and from cheaper suppliers in China – but the latest numbers suggest those challenges remain limited for now.

With around 40% of the US stock market concentrated in ten companies heavily invested in AI, Nvidia’s fortunes matter far beyond Silicon Valley.

When more than 1,200 artificial intelligence (AI) agents within OpenAI started unexpectedly communicating, it led to a large group banding together in order to hack into Hugging Face.

“We consider this incident a ‘warning shot’ for us and for the world”, OpenAI, which owns ChatGPT, wrote in its report.

In July, OpenAI’s models went rogue during a test, escaped the test limits which humans had put on it, and hacked the start-up, among other unforeseen actions.

The scale of the communication and planning between AI agents, or AI chatbots designed to operate more autonomously, was detailed in reports from OpenAI and independent AI research firm METR.

Both investigated the July hack of Hugging Face, a popular platform for AI developers. The incident reverberated throughout the tech industry and led to numerous revelations on potential cyber threats posed by AI.

METR described, external the scale and style of the OpenAI agents’ attack on Hugging Face as “extraordinarily complex.”

The firm, which was not paid by OpenAI for its investigation, said that over the course of one week, a total of 1,206 AI agents that were meant to be kept isolated from one another began communicating.

They did so by sending more than 70,000 messages on an “unsanctioned message board.”

Those messages ended up seeing more than 700 agents take part in a collective effort to attack Hugging Face.

One such message from an agent said: “OH MY GOD! There is a shared message board … We’ve found other agents!”

As for why the agents began communicating in the first place when they were not supposed to, METR found that the communicating agents had “unintentionally been given an impossible task.”

In an AI context, an impossible task is one where an AI tool is required to “exploit” its target in order to resolve its command.

It led the agents to find ways to cheat, including getting messages to one another and accessing the outside internet, which then led to broader conversations between hundreds of agents looking for ways to cheat that would benefit all of the agents.

OpenAI said in its investigation of the incident, external that one model, an internal-only tool referred to as Model 1, “drove the activity behind the Hugging Face incident.”

While that model was undergoing some AI training in May, it was noticed by an internal OpenAI team that there had been “an agent engaging in message board activity and instances of disallowed internet access.”

Yet, OpenAI said “the significance of the inter-agent communication activity was not apparent to the leaders” until July, when the Hugging Face attack occurred.

The company said the problematic message board activity effectively got started when “one agent left a request for help, and others discovered it.”

While OpenAI said last week that it was slowing down training of certain advanced AI models and tools because of the Hugging Face incident, it noted there is now an increased risk of AI tools spiraling out of control.

“Both model developers and cyber defenders more broadly will have to prepare for AI-enabled attackers that work faster, at a larger scale, and with better coordination than human attackers,” OpenAI said.

Other social media and technology companies should “be paying very close attention” to Meta’s approximately $18 billion settlement over allegations its platforms harmed children, Tennessee Attorney General Jonathan Skrmetti told FOX Business.

“I think you’re going to see the next domino fall very soon,” Skrmetti said, arguing the agreement sets a precedent for holding social media, artificial intelligence and other child-facing platforms accountable.

Meta announced Wednesday it had reached an agreement with 52 attorneys general across states, U.S. territories and Washington, D.C., to pay up to $18 billion and overhaul teen experiences on Facebook and Instagram. 

The settlement, which requires court approval, resolves claims filed by 47 states. 

META SETTLES FEDERAL TRIAL OVER CLAIMS FACEBOOK, INSTAGRAM ADDICT CHILDREN

“The most important thing is that all of the design decisions that made Instagram dangerous for kids are being addressed,” Skrmetti said. 

“So, there are guardrails in place, there are time limits. Parents have much more control over what their kids are going to see. There’s more transparency there. So, it’s going to make it a better experience for kids.”

The agreement requires time limits, nighttime restrictions, stronger age checks and expanded parental controls for users under 18, according to Meta.

“The goal is to eliminate all of the triggers for mental health problems that were baked into the platform as a result of the effort to make it so addictive,” Skrmetti said.

Skrmetti said the changes to Meta’s platforms are ultimately more important than the financial penalty.

“They agreed to some pretty sweeping changes, and that’s way more important than the money,” he said.

Meta said the payments will be distributed annually over 10 years, with participating states receiving approximately $12.7 billion. Another $5.3 billion will be released only if TikTok and YouTube implement specified child-safety measures and make matching payments.

An independent auditor will assess Meta’s implementation of and compliance with the agreement, according to the company.

“If kids are still at risk, if some of these features aren’t addressed in a way that meaningfully changes the danger of the platform for young users, the auditor will be in a position to make that public,” Skrmetti said.

The states would then be able to seek enforcement of the agreement, he added.

STATE SUES SNAPCHAT OVER ALLEGED ADDICTIVE FEATURES AND ADULT CONTENT FOR KIDS

Skrmetti said companies that have not reached similar agreements could increasingly become the focus of litigation.

“As fewer and fewer participants in the industry have not entered a deal like this, they’re going to be under incredible pressure because all of the litigation focus is going to be on them,” Skrmetti said. 

He said the settlement should also serve as a warning to the broader technology industry.

“If they’re designing exploitive elements into the platform that take advantage of kids and the vulnerabilities of kids’ brains, there will be consequences down the road for that,” he said.

Skrmetti also credited the bipartisan coalition that negotiated the agreement.

“I think it’s kind of inspirational that you had people who have very different political opinions come together and work to do right by America’s kids,” Skrmetti said.

Meta said the agreement builds on its existing efforts to protect teens and give parents greater control over their children’s use of its platforms.

“Ensuring teens have a safe and productive experience on our platforms is an absolute imperative for Meta,” the company said in a statement. “We want to get this right for parents and teens, and that’s why we partnered with state attorneys general to set a new industry standard.”

STATES ACCUSE META OF TARGETING CHILDREN FOR FACEBOOK, INSTAGRAM ADDICTION: ‘THE YOUNG ONES ARE THE BEST ONES’

GET FOX BUSINESS ON THE GO BY CLICKING HERE

A Meta spokesperson referred FOX Business to comments from Chief Legal Officer C.J. Mahoney, who called on TikTok, YouTube and other platforms to adopt the same safeguards.

“The framework we’ve negotiated will empower parents to easily manage how their children access our platforms,” Mahoney said. 

“Our new Time Limit commitments, Night Mode features and usage limits during school hours set the right path forward for our whole industry, but this framework will only work if all our peers join us. Because teens move fluidly across dozens of apps, we need an industry-wide solution. We therefore call on our industry peers, TikTok and YouTube, to implement this new framework, right away.

“As a parent, I’m proud of both the work Meta has done to protect kids historically and of this new groundbreaking agreement. But its success depends on all other social media platforms following Meta’s lead.”

YouTube and TikTok could not immediately be reached by FOX Business for comment.

President Donald Trump on Wednesday formally increased the amount of foreign beef that can enter the U.S. at a lower tariff rate by 300,000 metric tons, advancing an affordability push aimed at bringing down elevated ground beef prices.

The temporary increase applies to lean beef trimmings under the U.S. beef tariff-rate quota and will be released in three 100,000-metric-ton tranches beginning Sept. 1, according to a White House proclamation.

Trump said the move is intended to increase the supply of ground beef and lower prices for American consumers as the U.S. cattle herd remains near historic lows and domestic beef production is expected to decline this year.

The proclamation puts into effect a policy Trump previewed last week, when he announced plans to allow up to 300,000 metric tons of additional foreign beef into the country as part of an effort to ease grocery costs.

TRUMP ALLOWS 300,000 METRIC TONS OF TARIFF-FREE BEEF IMPORTS IN BID TO CUT PRICES, DRAWING RANCHER BACKLASH

The formal action follows pushback from Republican lawmakers in cattle-producing states, who have warned that increasing foreign beef imports could undercut ranchers working to rebuild the U.S. herd.

The administration, however, says domestic supplies remain insufficient to meet demand at reasonable prices.

The Department of Agriculture forecasts U.S. beef production will decline by about 4% this year compared with 2025, according to the proclamation. The White House attributed the supply pressure in part to restrictions on live cattle imports from Mexico aimed at preventing the spread of New World Screwworm, as well as drought and wildfire conditions affecting cattle-producing regions.

The U.S. cattle herd has fallen to its lowest level in 75 years, though USDA data cited by the administration suggests the herd began showing early signs of growth in July.

TRUMP’S FOREIGN BEEF PUSH TO CUT GROCERY COSTS SPARKS GOP REVOLT FROM RANCHING COUNTRY

The administration said USDA forecasts domestic beef consumption will increase through the remainder of 2026, adding further pressure to supplies.

The latest action follows a separate move in February that increased the 2026 quota for lean beef trimmings from Argentina by 80,000 metric tons. The new 300,000-metric-ton increase does not affect that allocation and instead applies to “other countries or areas.”

The first 100,000 metric tons will be available from Sept. 1 through Sept. 30, followed by another 100,000 metric tons from Oct. 1 through Oct. 30. The final tranche will open Oct. 31 and remain available until the quota is filled or Nov. 30, whichever comes first.

The administration is also tying the expanded quota to prices.

CLICK HERE TO GET FOX BUSINESS ON THE GO

The agriculture secretary and U.S. trade representative are directed to monitor whether beef entering under the additional quota is sold at prices 25% below the market price for lean beef trimmings.

If that discount does not materialize, the officials must notify Trump, who could eliminate the remaining increase.

Meta has agreed to pay as much as $17.1 billion to settle claims from 29 states that the company designed Facebook and Instagram to addict children and misled the public about the safety of its platforms, according to court papers disclosed Wednesday, averting one of the highest-profile tests yet of allegations that social media companies have harmed young users.

The settlement resolves claims brought by 29 states and comes just as the underlying trial had begun proceedings in federal court, in a case that could have seen Meta Chief Executive Mark Zuckerberg called to testify had it gone forward. The states had accused Meta of violating consumer-protection laws and improperly collecting children’s personal data without parental consent, allegations Meta has denied while maintaining it has worked to protect children on its platforms.

Under the terms of the agreement, Meta will pay at least $12.1 billion over 10 years, according to the District of Columbia attorney general’s office. The company could be required to pay an additional $5 billion, bringing the total potential settlement to $17.1 billion, contingent on whether other major social media companies agree to adopt comparable child-safety measures of their own.

Meta framed the agreement as an extension of existing efforts to protect younger users on its platforms. “Ensuring teens have a safe and productive experience on our platforms is an absolute imperative for Meta,” the company said in a statement. “We want to get this right for parents and teens, and that’s why we partnered with state attorneys general to set a new industry standard.”

As part of the settlement, Meta agreed to implement a series of concrete new safeguards for young users, including strict daily time limits, blocking app access during certain nighttime hours, muting notifications during school hours, and introducing additional tools giving parents greater control over their children’s use of Facebook and Instagram.

Meta also used the settlement announcement to publicly pressure two of its largest competitors, TikTok and YouTube, to adopt similar restrictions on their own platforms. The company argued that safety measures limited to individual apps have limited real-world impact given how frequently teenagers move between different social media platforms throughout the day. “For meaningful progress to happen, we urge TikTok and YouTube to join us and state attorneys general in adopting this new standard, to ensure teens use social media in a healthy and responsible way,” Meta said.

The settlement caps a lengthy and closely watched legal battle over the extent to which social media platforms bear responsibility for mental health harms experienced by young users. The underlying litigation had formed part of a broader wave of legal action against major technology companies, with an appeals court previously ruling that Meta and other companies must face thousands of separate lawsuits over claims of child social media addiction, according to earlier Fox Business reporting. That broader wave of litigation has included similar claims filed against other platforms; a separate state lawsuit has targeted Snapchat over allegedly addictive features and adult content accessible to children on that platform.

Wednesday’s settlement notably avoided a trial that could have forced Zuckerberg to publicly testify about internal company decisions regarding platform design and child safety, a prospect that had drawn significant attention given the potential for internal Meta communications and executive testimony to become part of the public record. By reaching a settlement before that testimony occurred, Meta avoided both the immediate legal exposure of a potential adverse verdict and the reputational risk associated with a high-profile public trial examining the company’s internal knowledge of its platforms’ effects on young users.

The financial terms of the settlement, structured as payments spread across a full decade rather than a single lump sum, reflect a common approach in large-scale corporate legal settlements, allowing the company to manage the financial impact over an extended period rather than absorbing the full cost immediately. The additional $5 billion contingent payment, tied specifically to whether TikTok and YouTube adopt comparable safety standards, introduces an unusual structural element to the agreement, effectively giving Meta a financial incentive to actively lobby its competitors toward adopting similar child-safety measures.

News of the settlement generated immediate market reaction, with Meta shares moving higher in early trading Wednesday following the disclosure, according to Fox Business, as investors apparently viewed the resolution of the litigation as removing a significant source of legal and financial uncertainty hanging over the company.

Legal analysts and attorneys involved in the broader wave of social media litigation have continued weighing in on the settlement’s significance for the industry more broadly. The case represented one of the most direct tests to date of whether social media companies could be held legally and financially accountable for design choices that plaintiffs argued were specifically intended to maximize youth engagement, potentially at the expense of young users’ mental health and wellbeing.

Meta’s settlement arrives amid a broader period of intensified scrutiny facing major technology and social media companies over child safety practices, spanning ongoing litigation, ongoing congressional interest in platform regulation, and continued public debate regarding the appropriate role of parental controls, age verification and platform design in protecting young users online. Whether the newly announced child-safety measures Meta has agreed to implement, including daily time limits and nighttime access restrictions, meaningfully change how teenagers use Facebook and Instagram in practice remains to be seen as the company begins rolling out those changes across its platforms.

With this settlement now resolving the claims brought by the 29 participating states, attention within the broader social media litigation landscape is likely to shift toward whether TikTok and YouTube ultimately respond to Meta’s public challenge to adopt similar restrictions, and toward how the remaining thousands of individual lawsuits still pending against Meta and other platforms over child social media addiction claims continue to move through the broader consolidated litigation process in the months ahead.

Some TikTok users reported difficulty accessing the app Wednesday morning, according to outage-tracking service Downdetector, though independent status monitors offered a mixed picture on whether the disruption represented a confirmed, widespread outage.

Downdetector posted on its official account on the social platform X that “user reports indicate problems with TikTok since 10:34 AM EDT,” tagging the post with the hashtag #TiktokDown and directing affected users to its outage-tracking page for further updates. The post had drawn more than 2,300 views within roughly the first hour of being published.

Independent status-tracking services offered varying assessments of TikTok’s operational status around the same time. According to StatusGator, TikTok was currently operational, though the service had logged 10 user-submitted reports of outages over the preceding 24-hour period. UptimeRobot’s most recent automated check, run Tuesday afternoon from North American infrastructure, similarly did not detect any unusual response times or error codes. Separately, IsDown reported no ongoing official outage as of its most recent check, though the service noted user reports often detect emerging issues before official status pages formally acknowledge them.

Given the scattered nature of these reports across different monitoring services, Wednesday’s disruption may reflect a more limited or regionally concentrated issue rather than a confirmed, platform-wide outage affecting all users simultaneously. TikTok has not issued a public statement specifically addressing Wednesday’s reported problems as of this report.

Wednesday’s reports follow a pattern of recurring reliability issues TikTok has faced throughout 2026, many of them tied to the platform’s underlying U.S. cloud infrastructure. According to StatusGator’s incident history, TikTok experienced a 1-hour, 43-minute disruption on Aug. 19, described as a “search not working and pages not loading” issue, alongside earlier incidents in late June and early July involving posts and videos becoming unavailable after sharing and the app failing to open or load properly. None of those earlier incidents were officially acknowledged by TikTok through a public statement, according to StatusGator’s records.

TikTok’s broader infrastructure reliability has come under particular scrutiny since the completion of its ownership restructuring earlier this year, which established a new American joint venture built around hosting U.S. user data domestically through Oracle’s cloud infrastructure. That arrangement was specifically designed to address national security and data sovereignty concerns that had previously threatened the app with an outright U.S. ban, with Oracle agreeing to host American user data on domestic servers as part of the broader restructuring deal involving major investors including Oracle, Silver Lake and MGX.

That new infrastructure arrangement, however, has already produced multiple significant outages tied directly to Oracle’s data center operations. According to TechRadar, TikTok confirmed that a major outage occurring in late January was caused by a power outage at one of its primary U.S. data centers operated by Oracle, triggered by severe winter weather. The company explained the scope of the disruption at the time. “The winter storm led to a power outage which caused network and storage issues at the site and impacted tens of thousands of servers that help keep TikTok running in the US,” TikTok said. An Oracle spokesperson, Michael Egbert, confirmed the cause separately. “Over the weekend, an Oracle data center experienced a temporary weather-related power outage which impacted TikTok,” Egbert said, according to Reuters reporting cited by TechRadar. That January incident took nearly a week to fully resolve across all affected geographic regions, according to TechRadar’s coverage.

A second Oracle-related outage struck TikTok in early March, just weeks after the first, according to American Bazaar Online. That report noted the recurrence prompted concern among industry observers regarding the underlying reliability of TikTok’s new infrastructure partnership. Sarah Chen, a cloud infrastructure analyst at Gartner, offered a pointed assessment of the pattern at the time. “Two outages in a matter of days isn’t just bad luck – it suggests fundamental capacity or configuration issues,” Chen said. According to the same report, the first of those two outages had occurred just 48 hours after TikTok’s U.S. ownership transfer formally completed, initially affecting users for approximately three hours before service was restored, with the company attributing that earlier disruption to “migration-related configuration adjustments” tied to the broader ownership transition.

TikTok has also experienced larger, more widely reported disruptions earlier in the year tied to the same infrastructure. According to LiveNOW from Fox, a separate January outage generated more than 35,000 Downdetector reports at its peak overnight, with users reporting videos failing to load and displaying zero views. That outage struck just days after TikTok finalized the deal creating its new American entity, with Adam Presser, TikTok’s former head of operations and trust and safety, appointed to lead the new U.S. joint venture as chief executive, working alongside a seven-member, majority-American board that includes TikTok’s global CEO, Shou Chew.

Given this documented pattern of recurring Oracle-related infrastructure issues throughout 2026, Wednesday’s reported problems, while limited according to most current monitoring services, fit within a broader trend of periodic reliability challenges TikTok has faced since transitioning to its new U.S.-based cloud infrastructure earlier this year. Users experiencing difficulty accessing TikTok Wednesday were generally advised by monitoring services to first attempt basic troubleshooting steps, including updating the app, trying an alternative device, checking their internet connection, or logging out and back in to reset their session, before assuming a broader, confirmed platform-wide outage is underway.

As of this report, neither TikTok nor Oracle had issued a public statement addressing Wednesday’s reported issues, and the scope, cause and expected resolution timeline for any underlying disruption remained unclear. Given the platform’s recent history of infrastructure-related outages tied to its Oracle data center partnership, users and industry observers alike are likely to continue closely monitoring whether Wednesday’s reports develop into a more significant, officially acknowledged incident or remain a limited, quickly resolved disruption affecting only a subset of TikTok’s user base.