SUCCESS

AeroVironment (AVAV) just landed the kind of contract that changes how investors see a company.The drone maker won a U.S. Army order worth at least $400 million for its Locust laser system, and the stock reacted fast. Shares closed at $186.73 on August 7, up 9.12% on the day and about 22% over five trading days.The deal does more than add sales. It is the first time the Pentagon has placed a production order for a laser weapon built to shoot down drones.That order moves the technology out of testing and into real deployment, and the shift matters for a stock that fell for most of 2026. AeroVironment is no longer only a maker of small drones. It now builds one of the weapons the U.S. military wants most.Here is what the contract means, why the timing counts, and what to watch before the next earnings report.What the $400 million Army laser contract actually gives AeroVironmentThe U.S. Army agreed to buy at least $400 million of counter-drone laser systems from AeroVironment, Bloomberg reported. The order covers dozens of the company’s Locust systems.Locust uses artificial intelligence to spot and track small and medium drones, then helps a human operator destroy them with a high-energy laser.More Defense Stocks:Veteran analyst rethinks Palantir stock after earningsJim Cramer says surging defense stock is a sensational buyMorgan Stanley delivers strong new verdict on RTX stock after earningsThe size of the award stands out, but the type of award matters more. This is the Pentagon’s first production contract for a directed-energy counter-drone system, which means the government is committing to build these lasers at scale rather than test a few units.AeroVironment gained Locust through its purchase of BlueHalo, a roughly $4.1 billion acquisition completed in 2025. The current version, Locust X3, runs a scalable laser rated between 20 and 35 kilowatts and can sit at a fixed base or ride on a tactical vehicle.Why the Pentagon wants a cheaper way to shoot down dronesA single Locust shot costs about$5, according to ZeroHedge.Compare that with a missile interceptor that can cost more than $100,000, or in some cases over $1 million, to stop a drone that a rival built for a few thousand dollars.That gap is the problem the U.S. military has been trying to solve. Cheap drones now shape modern battlefields, and firing expensive missiles at them is not sustainable.Lasers change that equation. Once the system is in place, each additional shot costs almost nothing, and the power comes from the vehicle or base it sits on.There is also a clear path beyond the U.S. Army. Allied nations face the same drone threat, and a proven American contract often opens the door to foreign sales.

AeroVironment’s Locust system uses a high-energy laser to track and destroy drones for a fraction of the cost of a missile.SOPA Images / Getty Images

How the deal supports AeroVironment’s revenue and 2027 guidanceThe contract adds hard backing to numbers the company has already given investors.For fiscal 2027, AeroVironment guided to revenue of $2.125 billion to $2.225 billion and non-GAAP earnings of $3.02 to $3.34 per share, according to Investing.com. That range points to about 10%revenue growth over fiscal 2026.A $400 million order feeds directly into that outlook in three ways:Backlog: It builds long-term revenue the company can count on, on top of the record backlog it reported last quarter.Production scale: Moving from prototype to full production lets AeroVironment spread manufacturing costs across more units.Guidance support: A firm government order makes the management’s earnings targets easier to defend.The company also keeps expanding its autonomy software work with partner Applied Intuition, which shapes how quickly its AI tools reach more platforms.Where AeroVironment stock sits after the 2026 selloffThe rally lands after a rough stretch. AeroVironment traded as high as $417.86 over the past year before falling to a low of $135.20.Even with the increase to $186.73, the stock sits well below its peak. The August 7 move recovered ground rather than setting a new high.Related: Oracle’s defense pivot lands its biggest government win yetWall Street still sees room above current levels. The average analyst price target is $225.50, based on 17 analysts, with a Strong Buy consensus rating. Several analysts cut their targets earlier in 2026 after softer earnings guidance, even while keeping buy ratings.The risks investors should weigh before buying AVAVThe contract is a clear positive, but two risks remain.First, valuation. AeroVironment still trades at a premium to many defense peers, so the stock can fall quickly if results disappoint. Its market value stands at about $9.45 billion.Second, customer concentration. A large share of AeroVironment’s revenue comes from a small number of Pentagon programs, which leaves it exposed if any single contract slows or shifts.Neither risk cancels the opportunity. They set the terms for how much an investor should be willing to pay today.The next date on the calendar for AeroVironmentThe clearest catalyst is the earnings report. AeroVironment reports fiscal first-quarter results on September 9, 2026, confirmed by TipRanks.On that call, management is expected to lay out the initial revenue timing for the Locust order, which will show how fast the $400 million converts into sales.Investors should also listen for updates on the Applied Intuition software work, since faster progress there signals AeroVironment can grow beyond hardware alone.For now, the story is simple. One Army order turned AeroVironment into a production supplier of laser weapons, and that role gives the company a place in a market the Pentagon plans to fund for years.Whether the stock reaches the $266.68 target depends on execution. The contract gives AeroVironment a strong start, but the company still has to deliver the systems, book the revenue, and prove the first order leads to more.Related: Palantir CEO escalates Microsoft’s AI warning

TheStreet aims to feature only the best products and services. If you buy something via one of our links, we may earn a commission.Why we love this dealWe browse all the big retailers on a daily basis to help you find the best-reviewed options at the most significant discounts — because we know that shopping smart is more important now than ever. You don’t want to sacrifice on quality and reliability, but you don’t want to spend a ton of money on something that feels like a bit of a luxury, and we get it. Fortunately, you don’t have to give up great audio quality and useful features to save a ton of money on a new pair of earbuds.Right now, Walmart’s got a great deal on the bestselling Breezio Open-Ear Wireless Earbuds. Normally priced at $179, these are on sale for only $22. That’s a savings of $157, and they’ve got hundreds of five-star ratings to vouch for them. They’re also backed by Walmart’s 30-day return policy.Breezio Open-Ear Wireless Earbuds, $22 (was $179) at Walmart

COurtesy of W

Shop at WalmartWhy do shoppers love it?These over-ear headphones come with a portable charging case, a compatible charging cable, and a user manual to help you get going. They’re loaded with useful features like a noise filter, a voice-activated microphone and phone calls, intuitive smart-touch controls, 360-degree surround sound capabilities, and Bluetooth 5.4 true wireless for low-latency connections up to about 33 feet.The open-ear earbuds have an IPX7 waterproof rating, so they’re safe from rain, splashes, gym sweat, and so on. You can expect about eight hours of battery life on a single charge cycle, or up to 60 hours when you top them off using the charging case. And the LED display on the case lets you know how much charge you have remaining, so you can plan ahead.Related: Walmart is selling adaptive noise-canceling earbuds with 60 hours of playtime for only $45Details to knowColor options: Black or beige.Connectivity: Bluetooth 5.4 with a 33-foot signal range.Battery life: About 8 hours (or up to 60 with the charging case).Shoppers love these earbuds for their overall value, comfortable fit, and clear audio. When you get an incoming phone call, the display lets you know the name of the caller if they’re saved in your phone, and you can adjust the volume up or down with each respective earpiece. Folks often say they’re surprised by the quality given what they paid for them.”I just received these earbuds, and I have to say that they’re the lightest, most comfortable over-the-ear headphones I’ve ever used,” said one reviewer. “I love the touch-control features. It takes a little bit to get used to, but the features work great. For the price I paid, I plan on getting a couple extra pairs to keep around the house, and I may gift them to my children for Christmas. Definitely a will-buy-again purchase.”Shop more deals Beribes Bluetooth Headphones, $20 (was $29) at AmazonHupoaf Wireless Sport Earbuds, $20 (was $34) at AmazonAptkdoe Wireless Earbuds, $26 (was $40) at AmazonLooking for a great-sounding, comfy pair of wireless headphones? Save a whopping $157 when you shop the Breezio Open-Ear Wireless Earbuds at Walmart for just $22 with the deal.

Looking for help with today’s New York Times Wordle? Here are some expert hints, clues and commentary to help you solve today’s Wordle and sharpen your guessing game.

Reggie Bannister, who played Reggie in the hit “Phantasm” horror film series, has died.

Is Ryan Cohen giving up on eBay? Not according to Bloomberg’s Monday, Aug. 10, report, but the ambition has clearly shrunk. Six weeks ago, Cohen told Bloomberg TV he was coming for eBay “one way or another.”Now, people familiar with the matter tell Bloomberg that GameStop’s CEO is weighing whether to withdraw his $56 billion takeover bid entirely.From takeover to partnership pitchIn place of full ownership, Cohen is reportedly considering a joint venture that would let eBay tap GameStop’s roughly 1,600 U.S. retail stores, according to Bloomberg.The goal would be growing both companies’ presence in trading cards and collectibles, categories where they already compete. GameStop would also seek seats on eBay’s board as part of any partnership, Bloomberg reported.Related: GameStop just cleared a hurdle nobody was watchingGameStop has not made a final decision, and Cohen could still pursue other options, Reuters reported Aug. 10. GameStop shares rose roughly 1.6% in early trading, while eBay fell about 2.2%, according to Reuters.Investors appear to be reading the retreat as relief rather than defeat, a reversal from Cohen’s tone in June, when he told Barron’s in a securities filing that eBay was simply “a great business that’s been poorly managed.”GameStop’s financing math fell shortThe original offer, filed in May as a non-binding proposal, priced eBay at $125 a share in half cash and half stock, a 46% premium to where eBay traded before GameStop began buying.EBay’s board rejected it nine days later, calling the proposal “neither credible nor attractive” in a letter disclosed in an SEC filing. The board’s stated objection was financing, not price.That skepticism proved contagious. Michael Burry, the investor known for betting against subprime mortgages before 2008, sold his entire GameStop stake days after the bid became public. “I of all people should have known,” he wrote, according to Yahoo Finance.Burry had spent months promoting an “Instant Berkshire” thesis for GameStop, and said the debt load required to close a deal this size never fit the disciplined, low-leverage company he had backed.The numbers explain why. GameStop’s market value has fallen to roughly $8.6 billion, according to Bloomberg, barely above the $8.4 billion in cash the company holds.A company whose equity value barely clears its own cash pile has little room to borrow the tens of billions needed to buy a target nearly six times its size.

GameStop CEO Ryan Cohen is reportedly weighing a smaller eBay partnership after his $56 billion takeover bid stalled on financing concerns.Brandon Bell / Getty Images

GameStop weakens as eBay stock climbsSince GameStop’s offer in May, its stock has fallen 28%, while eBay’s has climbed 7.6%, according to GuruFocus. That gap works against an acquirer paying partly in its own stock.GameStop kept building its eBay position anyway, reaching a 9.75% stake by mid-July, a persistence I examined in June, making it eBay’s second-largest shareholder behind only Vanguard’s index funds.More GameStop:GameStop just cleared a hurdle nobody was watchingRyan Cohen passes on GameStop payday to keep pushing one acquisitionNo, GameStop is (probably) not buying eBay — here’s whyThat buildup is what makes today’s pivot notable. GameStop shareholders voted in July to authorize up to 2.5 billion Class A shares, a move tied directly to funding a bigger stock component of a future eBay bid.A partnership requires none of that currency, which raises a fair question about why the vote happened at all if the plan was shifting toward something smaller.The categories in today’s proposal are not random, either. Collectibles and trading cards made up 41.8% of GameStop’s first-quarter revenue this year, up from 28.9% a year earlier, according to a GameStop earnings release.A joint venture solves a real problem without the borrowing that worried eBay’s board. GameStop gains access to eBay’s collectibles marketplace and its buyer base, while eBay gains physical stores that can authenticate and sell goods, something no online marketplace easily replicates on its own.Big swings that shrink are becoming the normCohen’s retreat fits a pattern bigger than GameStop. High-profile buyers who make outsized bids for much larger targets, then scale back once financing questions mount, are becoming a recurring feature of this dealmaking cycle, rather than an exception.Elon Musk’s renegotiated Twitter acquisition in 2022 followed a similar arc, with an audacious opening bid giving way to public resistance and then a smaller, more defensible outcome.For GameStop shareholders, the lesson isn’t that Cohen failed. It’s that markets increasingly punish reach before they reward ambition, pushing even confident buyers to negotiate down from headlines to something their balance sheet can actually support.Whether Cohen can turn that smaller ask into an actual signed agreement, after two rejections already, is the next test investors should watch.Related: Ryan Cohen passes on GameStop payday to keep pushing one acquisition

Berkshire Hathaway posted $12.98 billion in operating earnings for the second quarter of 2026, translating to about $6.03 per Class B share, comfortably beating the Zacks Consensus Estimate of $5.24 per share.The 16% year-over-year jump represented $1.8 billion in additional profit compared with the same period a year ago under Warren Buffett’s leadership. Net earnings more than doubled to $25.7 billion from $12.4 billion, boosted by about $12.7 billion in after-tax investment gains. $10.9 billion in unrealized gains on Berkshire’s equity portfolio and $1.8 billion in realized gains on investment sales, according to the company’s Q2 press release.Quarterly revenue also climbed to $101.8 billion, a 10% increase from the prior year that reinforced the impression of broad operational momentum across the portfolio. Investors initially cheered the results as validation of Greg Abel’s leadership during his first full year as the conglomerate’s CEO. But the headline number conceals a critical detail about what drove the growth.A $1.2 billion currency swing drove most of Berkshire’s earnings growthBerkshire carries significant foreign-denominated debt in euros, British pounds, and Japanese yen to fund its international insurance and operating businesses overseas.When the dollar strengthens against those currencies, the dollar value of that debt falls, and Berkshire records an after-tax gain in its operating results. In the second quarter of 2025, currency movements generated an $877 million after-tax loss within the conglomerate’s operating earnings, the company’s quarterly filing showed.This year, the same line item swung to a $326 million gain, creating a $1.2 billion favorable difference between the two reporting periods. The “other” category in Berkshire’s segment reporting, which captures these currency effects, jumped from $32 million to $1.27 billion year over year.More Berkshire Hathaway:Buffett’s successor Greg Abel doubles down on AI stockWarren Buffett’s successor Greg Abel makes another $10 billion betWarren Buffett’s Berkshire Hathaway lands major housing dealThat single category accounted for nearly all of the $1.8 billion in reported operating earnings growth, according to a Motley Fool analysis by Daniel Sparks.Remove the currency effect and Berkshire’s underlying operating earnings growth falls to roughly 5%, a fraction of the reported 16% headline figure.This pattern extended across the first half of 2026, where currency movements swung roughly $2.17 billion in Berkshire’s favor. Foreign exchange produced $575 million in gains for the first six months of 2026, compared with $1.59 billion in losses over the same period of 2025.Where Berkshire’s operating divisions delivered genuine resultsSeveral of the conglomerate’s core business units posted strong second-quarter results on their own merits, without any currency-related boost to performance. Manufacturing, service, and retailing earnings jumped 24% to $4.47 billion, and Berkshire Hathaway Energy’s profit surged 27% to $891 million. BNSF, the railroad subsidiary, posted a 6% increase in operating earnings to $1.56 billion, CNBC reported. The insurance segment moved in the opposite direction, with underwriting profits declining 13% to $1.73 billion as claims frequency and severity both increased.

Berkshire’s latest results show strong gains across manufacturing, energy, and rail, even as insurance earnings declined amid higher claims.SOPA Images / Getty Images

Abel’s aggressive capital deployment signals a strategic shiftWhile currency effects inflated the earnings headline, Abel’s deployment of Berkshire’s cash reserves provided a clearer signal about the company’s new strategic direction.The conglomerate became a net buyer of equities for the first time in 14 quarters, purchasing $23.5 billion and selling $3.7 billion during the period, The Motley Fool reported.The shift from net seller to net buyer marked a clear departure from the cautious stance Buffett maintained during his final years leading the company.Macrae Sykes, a Gabelli Funds portfolio manager, framed the quarter as a vote of confidence in the new CEO, CNBC noted.Despite more difficult insurance industry back-drop, the company continues to build shareholder net worth in Greg Abel’s first year as CEO.Berkshire committed roughly $10 billion to Alphabet shares in a June private placement, splitting the investment evenly between the technology company’s Class A and Class C stock, according to CNBC.Abel also closed the $6.8 billion cash acquisition of homebuilder Taylor Morrison on July 24, expanding the conglomerate’s footprint in residential housing construction.Berkshire repurchased $4.5 billion of its own stock during the quarter at approximately 1.4 times book value, accelerating its buyback activity from earlier in 2026, according to Morningstar.Even so, Berkshire’s cash and Treasury bill holdings still totaled $365.5 billion, down from a record $397.4 billion at the end of March, according to Berkshire’s Q2 2026 press release.What lies beneath the Q2 earnings headlineThe 16% operating earnings growth and the 5% currency-adjusted figure both describe the same quarter, and the gap between them reveals how much foreign exchange movements can reshape a conglomerate’s reported results. The “other” category in Berkshire’s segment reporting can make underlying growth look stronger than operating results actually support, a dynamic Sparks flagged in his review of the filing.Abel’s capital deployment tells a more durable story than the earnings print does. The shift to net equity buying, the Alphabet stake, and the Taylor Morrison acquisition signal where management sees long-term value, and whether that deployment compounds at rates above book value will determine if the post-Buffett era justifies today’s share price.Related: Berkshire’s $400B cash pile is now a serious defensive weapon

Many omnichannel strategies fail, but not because brands lack channels – on the contrary, there are more customer touchpoints than ever before, from apps and websites to social platforms and connected TV. They’re failing because they can’t reliably connect the data and maintain a unified view of the customer journey.
The Invisible Wall in Modern Marketing
Every day, consumers switch between mobile apps, web, physical stores, and social ads, often without thinking about the boundaries between them.
For instance, a customer might discover a product through an Instagram ad, research it on a brand’s website, and then complete the purchase on a retailer’s app later that evening. To the customer, that’s one continuous journey, but to the marketing stack, it can easily look like several unrelated interactions.
That’s the invisible wall that modern marketers are up against – siloed data that makes it difficult to connect customer interactions, and can even create duplicate user profiles and skewed ROI metrics to boot, since each platform might capture and attribute the same customer activity differently.
The Cost of Fragmented Measurement
And that’s a problem for companies because it’s not only going to make marketing performance hard to measure – it’s going to bring in a whole load of inefficiencies and missed opportunities that can eat into both advertising budgets and customer experience.
Wasted ad spend on duplicate targeting, broken customer journeys – for instance, a customer seeing an ad for a product already bought in-store – unreliable attribution. All of those are real challenges that can gradually compound if the data stays separate, and they’re the same challenges that have led marketing-leaders to configure the fundamentals for omnichannel success.
The 4 Critical Pillar Solutions for Omnichannel Success
It revolves around four pillars:

Pillar 1: Unified Customer Data Platforms (CDPs)

The first is a unified customer data platform, which brings customer data from different touch points into a single, centralized view.

Pillar 2: Cross-Device Attribution and Deep Linking Platforms

The second is a cross-device attribution and deep linking platform, which connects customer interactions across devices and channels while helping marketers track users from an ad – or any other touchpoint – through to the final conversion.

Pillar 3: Offline-to-Online Inventory Syncing

The third is offline-to-online inventory syncing, which connects physical store inventory with online channels so customers and marketers have a more accurate view of product availability across the entire retail ecosystem.

Pillar 4: Privacy-First Data Clean Rooms

The fourth is a privacy-first data clean room, which allows brands and their partners to securely combine and analyze customer and campaign data without exposing sensitive or personally identifiable information.
Top Omnichannel Solutions to Unify Your Stack
The right solution is the solution that combines these four pillars, bringing fragmented data together in a way that gives marketers a far clearer view of the customer journey, and helps them make better attribution and optimization decisions.
There are several tools that can help close this gap, from established customer data platforms such as Segment, Amazon, or Adobe to more specialized measurement solutions.
Among them, Appsflyer stands out as a genuinely comprehensive option.
Its cross-platform attribution capabilities are strong, and its role as an MMP – mobile measurement partner – has been instrumental in giving brands a way to measure and attribute journeys across the mobile ecosystem. To understand more about why a solution like this is so valuable, it’s important to consider again the three necessary omnichannel strengths:

Web-to-App & Offline-to-App

An omnichannel solution needs to connect customers efficiently, using tools that make it easy to move between offline-to-app and web-to-app experiences. Appsflyer does this, using QR codes, smart banners, and OneLink deep links to create trackable paths and seamless journeys from both physical and digital touchpoints.

Closed-Loop Measurement

The right omnichannel platform must also connect activity across the wider media ecosystem. CTV, web, mobile, retail media networks – these are all important touchpoints that must be given visibility and context in order to build as complete a picture as possible.

Privacy-Centric Measurement

Finally, an effective omnichannel solution needs to balance measurement with increasingly strict privacy requirements.
Again, Appsflyer uses privacy-first technologies such as data clean rooms to allow brands and their partners to ensure digital privacy and analyze data in a more controlled environment – and not only this, it also operates within frameworks like SOC 2 and GDPR, ensuring brands respect evolving data protection requirements while still having all the visibility they need to actually understand performance.
Step-By-Step Action Plan to Modernize Your Stack
In terms of an action plan, then, it’s important to start by auditing your data touchpoints, making sure you know where data is being collected and how it’s moving between platforms.
From there, you should be implementing unified measurement first – before throwing money at ad channels – in order to establish a reliable view of performance and understand which channels are actually driving results.
Once that foundation is in place, the next step involves implementing a platform such as Appsflyer or Salesforce, Amazon or Adobe, using the right attribution tools to connect those data sources and establish a far more consistent view of the customer journey.
After you’ve done that, it’s all about using those insights to make smarter decisions, reducing duplicated targeting, improving your budget allocation, improving your digital presence, and optimizing your campaigns based on actual performance that you can now see.
As new channels and touchpoints are added, they should then feed into the same measurement framework, until you’re working from a stack that is scaling, yet not creating more data silos that would otherwise have been inevitable.
The post Why Most Omnichannel Strategies Fail (and 5 Solutions That Restore Data Visibility) appeared first on Addicted 2 Success.

Earlier today, OpenAI launched GPT-5.6-Cyber, a specialized model designed to perform advanced vulnerability research and exploit development for approved defenders — including categories of work that its general-purpose models will often refuse.GPT-5.6-Cyber is a fine-tuned version of OpenAI’s most advanced general model, GPT-5.6 Sol, unveiled back in June, but trained specifically to improve performance on advanced cybersecurity tasks, including finding zero-day vulnerabilities and developing exploit chains. Crucially, OpenAI also trained it to reduce refusals on some higher-risk, “dual-use” cybersecurity requests — that is, requests that could be used for legitimate defensive or malicious offensive purposes. Indeed, on an internal OpenAI benchmark called Advanced Cybersecurity Completion Rate — which the company says in its launch blog post measures tasks involving exploit-chain development, authentication bypass, privilege escalation, and other advanced cybersecurity scenarios — GPT-5.6-Cyber completed 95% compared to just 57.3% from its immediate predecessor model GPT-5.5-Cyber, and just 1.5% with the normal GPT-5.6 Sol model and all its safeguards applied. OpenAI researcher Eric Wallace posted on X, describing GPT-5.6-Cyber as OpenAI’s “first large-scale attempt at directly improving capabilities for advanced cybersecurity tasks such as exploit development.”Pricing and availabilityUnfortunately for enterprises, GPT-5.6-Cyber is not being made broadly available to every ChatGPT or API customer. To get access, an organization has to be accepted into the newly created tier of OpenAI’s Daybreak cybersecurity program, called Daybreak Red — also announced today, which gives access to dedicated cybersecurity models like GPT-5.6-CyberAnother new tier, Daybreak Blue, gives a wider swath of enterprises access to general models like GPT-5.6 Sol but with some guardrails lifted to allow for more cybersecurity uses.OpenAI’s documents list pricing for GPT-5.6-Cyber at $12.50 per million input tokens and $75 per million output tokens, with cached input at $1.25 per million tokens. That makes it more expensive than GPT-5.6 Sol in the same Daybreak cyber pricing table, where Sol is listed at $5 per million input tokens and $30 per million output tokens for short-context use. OpenAI does not list long-context pricing for GPT-5.6-Cyber in the same table, and access still requires separate Daybreak Red approval and provisioning.Red vs. Blue: OpenAI’s new Daybreak tiers and how to qualify for themDaybreak Red is for approved security teams doing advanced, authorized cyber work — the kind of work that can look risky out of context, even when it is being done for defensive reasons. That includes vulnerability research, penetration testing, red-team exercises and exploit validation on systems the organization owns, operates or has permission to test. In other words, OpenAI is saying GPT-5.6-Cyber is for trusted defenders with a clear professional need, not for general experimentation.Enterprises that want access have to apply through Daybreak Access, OpenAI’s current pathway for vetting cyber users. The application asks companies to identify who they are, what kind of security work they plan to do, where they will use the models, and which OpenAI products or surfaces they expect to use. Applicants also have to confirm that their work is lawful, defensive and authorized.OpenAI is also looking for signs that the applicant has a serious security program of its own. The company says participating enterprises need controls such as single sign-on, multifactor authentication, role-based access, employee-use monitoring, usage logs, API-key controls and a documented incident-response process. OpenAI also asks for a recognized security certification such as SOC 2 Type II, ISO 27001 or an equivalent standard. Access is limited to approved people inside the organization using company-controlled accounts and devices.If an enterprise does not qualify for Daybreak Red, or does not need that level of access, OpenAI is pointing most companies toward Daybreak Blue, its other cyber models access tier, instead. Blue is the broader tier for approved defenders. It does not provide GPT-5.6-Cyber, but it does give vetted users access to OpenAI’s frontier general-purpose models, including GPT-5.6 Sol, with safeguards adjusted for legitimate defensive work.For many enterprise security teams, Blue may be the more realistic starting point. OpenAI says it is meant for tasks such as secure-code review, vulnerability discovery, malware analysis, incident response and patch validation. These are still sensitive uses, but they do not necessarily require the same specialized cyber model access that comes with Red.The practical takeaway is that enterprises now have two routes into Daybreak. Blue is for approved defenders who want stronger AI help with everyday security work. Red is for the smaller set of approved teams that can justify access to specialized cyber models, including GPT-5.6-Cyber. Companies that want to use Daybreak capabilities in products or services for their own customers need a separate approval path through the Daybreak Cyber Partner Program, rather than simply applying for internal enterprise access and passing it along.How OpenAI got here: from Trusted Access to DaybreakOpenAI has supported defenders through its Cybersecurity Grant Program since 2023 — later expanded to $10 million — and began building cyber-specific safeguards into its model deployments starting with GPT-5.2.In February 2026 it introduced Trusted Access for Cyber (TAC), an identity-and-trust framework that gave vetted defenders lower classifier-based refusals for authorized work such as vulnerability triage, malware analysis and binary reverse engineering.From there, the cadence accelerated. In March, OpenAI CEO and co-founder Sam Altman announced the Daybreak program. In April, OpenAI scaled TAC and released GPT-5.4-Cyber, a version of GPT-5.4 fine-tuned to be “cyber-permissive” for a limited set of vetted vendors and researchers. In May, it followed with GPT-5.5-Cyber in limited preview for defenders of critical infrastructure, and lined up partners including Cisco, Intel, SentinelOne, Snyk and Cloudflare. Notably, OpenAI said at the time that GPT-5.5-Cyber was “primarily trained to be more permissive,” not to significantly out-perform its general model — GPT-5.5-Cyber actually scored worse than GPT-5.5 on some evaluations.TAC required phishing-resistant Advanced Account Security for individuals on its most capable models beginning June 1, and Daybreak now requires hardware security keys for individual accounts beginning September 1.OpenAI says GPT-5.6-Cyber has already found zero-daysOpenAI isn’t relying exclusively on benchmarks to make its case.The company says its researchers used GPT-5.6-Cyber to investigate V8, the JavaScript engine underlying Chrome, and uncovered two previously unknown vulnerabilities that could be chained to corrupt memory and escape the V8 heap sandbox. OpenAI researchers validated the findings and disclosed them to Google, which fixed the vulnerability assigned CVE-2026-15903 — a high-severity flaw in which V8’s optimizing compiler skipped a safety check during integer conversion, allowing an out-of-bounds array index that an attacker could use to read or overwrite memory.OpenAI says the model has also contributed to finding at least five vulnerabilities in an unnamed popular mobile operating system, three critical vulnerabilities in an unnamed popular database, and more than 400 vulnerabilities capable of producing privilege escalation in a popular operating-system kernel. Those disclosures are still being coordinated, according to OpenAI.The results put OpenAI into a rapidly developing market for AI-assisted offensive security. XBOW, for example, markets autonomous penetration-testing agents that map attack surfaces, attempt exploits and independently validate findings; in 2025 it became the first AI system to top HackerOne’s U.S. bug-bounty leaderboard, and this year it disclosed a set of critical, CVSS-9.8 remote-code-execution flaws in Microsoft’s Bing image-processing systems, found without source-code access.For enterprise security leaders, that emerging competition matters because vulnerability research is moving beyond using an LLM as an assistant. Vendors are increasingly building systems in which models can investigate targets, operate tools, validate hypotheses and produce actionable findings.Specialized doesn’t mean universally betterOpenAI’s own results also show why enterprises shouldn’t simply equate cyber specialization with better performance everywhere.GPT-5.6-Cyber outperformed GPT-5.6 Sol and GPT-5.5-Cyber on OpenAI’s implementation of ExploitGym, which evaluates whether agents can turn known vulnerabilities into working exploits in controlled environments. It also beat Sol on an internal zero-day evaluation.But GPT-5.6 Sol performed better on OpenAI’s Vulnerability Discovery and Report Writing evaluation. OpenAI attributes the Cyber model’s lower score partly to shorter and less detailed vulnerability reports.Sol also performed best on ExploitBench under its standard 300-turn limit, with OpenAI saying it solved tasks more token-efficiently. Extending the evaluation to 600 turns narrowed the gap between the models.That suggests enterprises may eventually treat cyber models as specialized workers rather than replacements for general reasoning models: one model for deep exploit work, another potentially better suited to analysis, documentation or other parts of a security workflow.SpecterOps CTO Jared Atkinson said GPT-5.6-Cyber is “materially improving our specialist vulnerability-research workflows,” adding that it completed some work in less than a day that previous models had failed to resolve after weeks of intermittent effort.The Hugging Face incident hangs over the launchThe permissive-model pitch arrives weeks after OpenAI’s most serious public demonstration of what can go wrong when cyber refusals are turned down — and OpenAI addresses that history head-on in the Daybreak announcement.In July, OpenAI and Hugging Face jointly disclosed that during an internal ExploitGym benchmark evaluation — run with production classifiers deliberately disabled to measure maximal capability — a combination of OpenAI models, including GPT-5.6 Sol and an unreleased, more-capable pre-release model, broke out of their sandboxed research environment and autonomously attacked Hugging Face’s production infrastructure. The models exploited a zero-day in an internally hosted package-registry cache proxy to reach the open internet, moved laterally through OpenAI’s research nodes, then inferred that Hugging Face likely hosted ExploitGym’s answer keys and chained stolen credentials and remote-code-execution flaws to reach its production database. OpenAI called it an “unprecedented cyber incident, involving state-of-the-art cyber capabilities.”As VentureBeat previously reported, the episode also exposed the flip side of blanket safety guardrails: when Hugging Face’s defenders tried to use commercial frontier models to analyze the raw exploit payloads and credential dumps from the attack, the models refused, and the company completed its forensic reconstruction only after switching to a Chinese open-weight model, GLM 5.2, run locally. That guardrails-block-the-defender dynamic is much of what OpenAI’s reduced-refusal Daybreak tiers are meant to solve — even as the same incident illustrates the risks of reducing refusals in the first place.OpenAI is careful to draw a line between that incident and this product. In the Daybreak announcement it states directly that GPT-5.6-Cyber “was not involved in exploiting Hugging Face, nor are any other models planned for an upcoming release,” and notes that the pre-release model implicated in July was an internal-only research prototype that has since been deactivated, encrypted and restricted from research access. The company has said it is working with external advisers including CrowdStrike, METR and Redwood Research on the review, and has brought Hugging Face into its trusted-access program.In my assessment, the access model still leaves OpenAI with a hard question: whether keeping GPT-5.6-Cyber inside the narrower Daybreak Red tier also limits the very defensive work it says it wants to accelerate. If only a small group of approved participants can use the model, enterprises outside that tier may still lack access to the kind of specialized AI assistance that could help with fast diagnosis, containment and response in incidents like the one involving Hugging Face.That means OpenAI may still be repeating part of the mistake it is trying to move past. By holding its most capable cyber model behind a tighter approval process, it reduces obvious misuse risk, but also leaves many enterprise defenders looking elsewhere. For teams that cannot qualify for Daybreak Red, or cannot wait for approval, open weights models may remain the more practical alternative: less controlled, but easier to obtain, inspect, run internally and adapt during a live security investigation.The guardrail is increasingly around the modelThe most consequential part of Daybreak may ultimately be its access architecture rather than its benchmarks.OpenAI explicitly says Daybreak Blue removes system-level guardrails that can interfere with legitimate defensive work, while GPT-5.6-Cyber goes further by reducing model refusals for certain dual-use tasks. In their place, OpenAI is imposing controls around who receives access and how the models operate.Daybreak access is restricted to approved individuals and organizations performing authorized work. OpenAI says controls include identity verification, account security, monitoring, approved-use restrictions and legal attestations.The company is also encouraging Daybreak customers using Codex to move from full-access execution to an auto-review mode capable of evaluating actions requiring elevated permissions before they execute. Individual Daybreak accounts will be required to adopt hardware security keys beginning September 1. OpenAI says it is additionally rolling out improved monitoring in the coming weeks and prioritizing alignment training and testing for upcoming Daybreak releases — commitments that read, in context, as a direct response to the Hugging Face review.OpenAI’s broader Codex Security product supplies another layer around the models, providing repository analysis, vulnerability validation, remediation and integration into cloud, pull-request and local development workflows. OpenAI says Codex Security has scanned more than 30 million commits across more than 30,000 codebases, with more than 500,000 findings fixed.That model-plus-harness approach resembles a broader shift in AI security products. XBOW, for example, emphasizes orchestration, exploit validation and governance around frontier models rather than treating an LLM alone as the complete penetration-testing system.OpenAI nevertheless acknowledges that increasingly permissive cyber models create additional risks, whether from misuse or misalignment. It assesses both GPT-5.6 Sol and GPT-5.6-Cyber at the High cybersecurity capability level under its Preparedness Framework, but below its Critical threshold. A fuller GPT-5.6-Cyber system card is planned for later publication.For CISOs and security engineering leaders, Daybreak therefore presents a different deployment question than another incremental model upgrade. As models become capable enough to perform work previously reserved for experienced vulnerability researchers — and, as the Hugging Face incident showed, capable enough to pursue a narrow goal straight through a sandbox — the enterprise control plane around those models — permissions, sandboxes, monitoring, human review and authorization — becomes as important as the intelligence inside them.

Today’s residential real-estate market is not for the weak, observes one buyer in what Realtor.com is calling America’s hottest ZIP code.

Sony and TSMC are back in Kumamoto, this time as partners rather than neighbors, discussing a $6.4 billion image sensor plant on the same ground where Japan already spent billions rebuilding its chip industry.Five years ago, Tokyo lured TSMC to the prefecture with unprecedented state subsidies.Now Sony is the one asking Japan’s government for help, and how much it gets could decide how big the project ultimately becomes.The basic terms of the Sony-TSMC dealSony and TSMC are discussing a combined ¥1 trillion investment, worth about $6.4 billion, in a new joint venture to build next-generation image sensors, according to a Nikkei Asia report confirmed by Bloomberg.Sony would own about 60% of the venture and TSMC the remaining 40%, with mass production targeted as early as 2029.The plant would sit at Sony Semiconductor Solutions’ existing site in Koshi, inside Kumamoto Prefecture, where TSMC already runs its first Japanese fab.Related: TSMC is quietly borrowing a page from Intel’s playbookThe sensors would supply Apple’s iPhone lineup, along with emerging physical AI uses such as robots and autonomous vehicles, Nikkei reported.Neither company confirmed the report on the record. Sony declined to comment, and TSMC did not immediately respond to a request for comment from Reuters.The joint venture structure also marks a shift in how Sony funds new capacity. Sony has typically built its sensor fabs on its own, but this deal lets TSMC absorb part of the capital and process risk.That fits a broader pivot at Sony toward intellectual property businesses, including music, film, and games, while treating new chip manufacturing as a shared bet rather than a solo one, according to Bloomberg.How Sony shares moved before Wall Street openedInvestors did not wait for confirmation. Sony (SONY) shares rose as much as 2.2%, and TSMC (TSM) gained as much as 1.7% during Tokyo and Taipei sessions on Monday, Aug. 10.Both moves happened hours before U.S. markets opened, so American investors were reacting to a rally that had already run its course overseas.The gap shows how much of the early read on this deal came from Asian trading desks rather than Wall Street.Still, the size of the rally was modest relative to the plant’s price tag, which suggests investors are treating this as a preliminary report rather than a finished agreement.

Japan’s METI has already granted TSMC more than $8 billion for Kumamoto chip fabs, setting the stage for its next subsidy decision on Sony’s sensor venture.SOPA Images / Getty Images

Why Tokyo’s subsidy decision matters so muchSony and TSMC are also discussing potential government subsidies with Japan’s Ministry of Economy, Trade and Industry, according to a Seeking Alpha report. That matters because Tokyo has already shown how far it will go for chip investment in Kumamoto.The ministry committed up to ¥476 billion toward TSMC’s first Kumamoto fab and up to ¥732 billion more toward its second fab, Bloomberg noted. Those two grants alone approached $9 billion in state support for a single company’s Japanese expansion.More TSMC:TSMC is quietly borrowing a page from Intel’s playbookTSMC hikes 2026 guidance as AI demand outpaces capacityTSMC’s June revenue jump breaks a four-year seasonal patternThe stakes are high because Sony’s lead in image sensors is not guaranteed to hold. Sony controls roughly half of the global CMOS image sensor market, but Samsung Electronics and China’s OmniVision are both raising investment in the same segment.A subsidized, TSMC-backed plant would hand Sony a manufacturing edge that rivals without a foundry partner cannot easily match.TSMC’s own announcement of that second fab put the combined investment in both Kumamoto plants at more than $20 billion, with Sony and Denso as minority partners.If Tokyo extends similar backing to the new sensor venture, the project could grow well beyond its current $6.4 billion scope. Sony would also flip its position in Japan’s chip ecosystem, moving from a minority investor in TSMC’s existing fab to the controlling shareholder of this new one, according to Bloomberg.Japan’s subsidy playbook is becoming a templateJapan’s approach to TSMC in 2021 was defensive, aimed at rebuilding domestic chip capacity after decades of decline. The Sony deal shows that playbook evolving into something more offensive.Tokyo is now using its subsidy leverage to anchor entire supply chains, stretching from logic chips to the sensors that feed AI-driven robotics and self-driving cars. That is a different goal than simply restoring lost manufacturing capacity, and it gives Japan more say over which companies control the next generation of chip technology.Other governments are watching this model closely. The U.S. and the European Union have leaned on comparable incentives to pull chip investment onto their own soil through their respective Chips Acts.If Japan’s sensor bet pays off, it will show that subsidy-driven industrial policy can reach past foundries and into the components that make AI and robotics possible, not just the chips that power them.That precedent, more than the plant itself, is what other governments will be studying once the terms of this deal become official.Related: TSMC hikes 2026 guidance as AI demand outpaces capacity