15.5 C
New York
Monday, March 31, 2025
Home Blog Page 4

ios – SwiftUI Menu in Checklist solely the Textual content is tappable


I count on that UIKit may allow you to management the place of the menu extra exactly, however, sadly, you do not have a lot management with SwiftUI. I’m guessing, the Reminders app shouldn’t be applied utilizing SwiftUI.

For the case proven within the first block of code, you may improve the tappable space by making use of a body to the menu label and setting a content material form. That is just like what you had been doing within the potential answer you confirmed within the query. It may be carried out with out an HStack:

Menu {
    // ... buttons, as earlier than
} label: {
    Textual content("Add Query")
        .foregroundStyle(Coloration.white)
        .background(Coloration.black)
        .body(maxWidth: .infinity, alignment: .main)
        .contentShape(Rectangle())
}

Nevertheless, while you faucet the row, the menu all the time aligns to the center of the row, even while you faucet on the textual content (as you additionally defined within the query):

screenshot

Listed here are some potential workarounds, none of that are excellent, however they may assist:


1. Experiment with minWidth

In my exams on an iPhone 16 simulator, the purpose at which it switches from left-alignment to center-alignment is when the width of the label is larger than about 273 factors. That is about two-thirds of the display screen width on an iPhone 16, nevertheless it may be totally different on different units. So if the body on the label is modified from maxWidth: .infinity to, say, minWidth: 270, this provides a bigger hit space with the menu nonetheless left-aligned. Nevertheless, the hit space won’t embrace the trailing space on the suitable of the row.

Textual content("Add Query")
    .foregroundStyle(Coloration.white)
    .background(Coloration.black)
    .body(minWidth: 270, alignment: .main) // 👈 modified
    .contentShape(Rectangle())

2. Offset the label with padding to compensate

As we now have seen, when the label has most width, the menu is centered. It seems that if the middle of the row is shifted, the menu strikes too.

The middle of the row might be shifted by making use of a optimistic x-offset to the label, then compensating with unfavorable main padding on the Menu.

  • Through the use of .offset as a substitute of main .padding for the label, the width accessible for the textual content doesn’t change. So an extended label will solely wrap if the total row width is simply too slim.
  • Through the use of unfavorable padding for the Menu, the hit space doesn’t transfer. If as a substitute a unfavorable offset is used, the right-side of the row is now not receptive to faucets.
personal let labelOffset: CGFloat = 20
Menu {
    // ... buttons, as earlier than
} label: {
    Textual content("Add Query")
        .foregroundStyle(Coloration.white)
        .background(Coloration.black)
        .body(maxWidth: .infinity, alignment: .main)
        .contentShape(Rectangle())
        .offset(x: labelOffset) // 👈 added
}
.padding(.main, -labelOffset) // 👈 added

When the dimensions of the offset is small, the menu strikes by half this measurement. For instance, if the offset is 20 (as above), the menu is proven 10 factors to the left of middle.

There appears to be a threshold, above which the menu alignment switches from the middle of the row to the left facet of the row. In my exams on an iPhone 16 simulator, the edge was round 40. This worth occurs to be the gap from the sting of the display screen to the row content material. When the checklist row insets are modified, the edge additionally modifications. So the edge in all probability is determined by the insets.

Rising the offset above this threshold makes no distinction. So to maneuver the menu to the left, the offset would not have to be an actual quantity, it simply must be greater than the edge.

Right here is the way it appears to be like utilizing an offset of fifty. It additionally appears to be like precisely the identical when you use an offset of 100:

Screenshot

To tremendous tune the positioning, it could be good if the menu may very well be moved to the suitable by about 12 factors, in order that it aligns with the row. Sadly, I could not discover a approach to do that😢


3. Add trailing padding to the menu

As type of a mixture of the 2 workarounds above, the menu may also be moved to the left facet by including trailing padding to the menu.

The quantity of padding that’s wanted corresponds once more to the edge offset mentioned above for workaround 2. So with the default checklist insets, the padding for an iPhone 16 must be at the very least 40. On an iPad, it must be bigger (100 works).

  • The hit space on the right-side of the row is lowered by the padding quantity, so it’s best to maintain the padding to the minimal essential.

  • Destructive padding might be added to the label to stop the textual content from wrapping, if wanted. Nevertheless, this does not assist to extend the hit space.

This variant offers higher menu alignment. So although the hit space excludes the width of the padding on the suitable facet of the row, it’s a higher answer than workaround 1:

Menu {
    // ... buttons, as earlier than
} label: {
    Textual content("Add Query")
        .foregroundStyle(Coloration.white)
        .background(Coloration.black)
        .body(maxWidth: .infinity, alignment: .main)
        .padding(.trailing, -40) // 👈 added
        .contentShape(Rectangle())
}
.padding(.trailing, 40) // 👈 added

Screenshot

The Rise of Smarter Robots: How LLMs Are Altering Embodied AI

0


For years, creating robots that may transfer, talk, and adapt like people has been a significant purpose in synthetic intelligence. Whereas vital progress has been made, growing robots able to adapting to new environments or studying new abilities has remained a fancy problem. Latest advances in massive language fashions (LLMs) are actually altering this. The AI methods, skilled on huge textual content information, are making robots smarter, extra versatile, and higher capable of work alongside people in real-world settings.

Understanding Embodied AI

Embodied AI refers to AI methods that exist in bodily kinds, reminiscent of robots, that may understand and work together with their setting. In contrast to conventional AI, which operates in digital areas, embodied AI permits machines to interact with the bodily world. Examples embody a robotic selecting up a cup, a drone avoiding obstacles, or a robotic arm assembling components in a manufacturing facility. These actions require AI methods to interpret sensory inputs like sight, sound, and contact, and reply with exact actions in actual time.

The importance of embodied AI lies in its potential to bridge the hole between digital intelligence and real-world purposes. In manufacturing, it might probably enhance manufacturing effectivity; in healthcare, it may help surgeons or assist sufferers; and in properties, it may carry out duties like cleansing or cooking. Embodied AI permits machines to finish duties requiring extra than simply computation, making them extra tangible and impactful throughout industries.

Historically, embodied AI methods have been restricted by inflexible programming, the place each motion wanted to be explicitly outlined. Early methods excelled at particular duties however failed at others. Trendy embodied AI, nevertheless, focuses on adaptability—permitting methods to study from expertise and act autonomously. This shift has been pushed by advances in sensors, computing energy, and algorithms. The combination of LLMs is starting to redefine what embodied AI can obtain, making robots extra able to studying and adapting.

The Position of Giant Language Fashions

LLMs, reminiscent of GPT, are AI methods skilled on massive datasets of textual content, enabling them to know and produce human language. Initially, these fashions have been used for duties like writing and answering questions, however they’re now evolving into methods able to multimodal communication, reasoning, planning, and problem-solving. This evolution of LLMs is enabling engineers to evolve embodied AI past performing some repetitive duties.

A key benefit of LLMs is their potential to enhance pure language interplay with robots. For instance, while you inform a robotic, “Please fetch me a glass of water,” the LLM permits the robotic to know the intent behind the request, determine the objects concerned, and plan the required steps. This potential to course of verbal or written directions makes robots extra user-friendly and simpler to work together with, even for these with out technical experience.

Past communication, LLMs can help with decision-making and planning. For instance, when navigating by means of a room stuffed with obstacles or stacking bins, an LLM can analyze information and recommend the very best plan of action. This potential to suppose forward and adapt in real-time is important for robots working in dynamic environments the place pre-programmed actions are inadequate.

LLMs may also assist robots to study. Historically, educating a robotic new duties required intensive programming or trial-and-error. Now, LLMs allow robots to study from language-based suggestions or previous experiences saved in textual content. For example, if a robotic struggles to open a jar, a human would possibly say, “Twist more durable subsequent time,” and the LLM helps the robotic alter its method. This suggestions loop refines the robotic’s abilities, enhancing its capabilities with out fixed human oversight.

Newest Developments

The mixture of LLMs and embodied AI isn’t just an idea—it’s taking place now. One vital breakthrough is utilizing LLMs to assist robots deal with advanced, multi-step duties. For instance, making a sandwich entails discovering substances, slicing bread, spreading butter, and extra. Latest research present that LLMs can break down such duties into smaller steps and alter plans based mostly on real-time suggestions, like if an ingredient is lacking. That is essential for purposes like family help or industrial processes the place flexibility is essential.

One other thrilling growth is multimodal integration, the place LLMs mix language with different sensory inputs, reminiscent of imaginative and prescient or contact. For instance, a robotic can see a purple ball, hear the command “decide up the purple one,” and use its LLM to attach the visible cue with the instruction. Initiatives like Google’s PaLM-E and OpenAI’s efforts present how robots can use multimodal information to determine objects, perceive spatial relationships, and carry out duties based mostly on built-in inputs.

These developments are resulting in real-world purposes. Corporations like Tesla are incorporating LLMs into their Optimus humanoid robots, aiming to help in factories or properties. Equally, LLM-powered robots are already working in hospitals and labs, following written directions and performing duties like fetching provides or conducting experiments.

Challenges and Concerns

Regardless of their potential, LLMs in embodied AI include challenges. One vital difficulty is making certain accuracy when translating language into motion. If a robotic misinterprets a command, the outcomes could possibly be problematic and even harmful. Researchers are engaged on integrating LLMs with methods focusing on motor management to enhance efficiency, however that is nonetheless an ongoing problem.

One other problem is the computational calls for of LLMs. These fashions require substantial processing energy, which will be tough to handle in real-time for robots with restricted {hardware}. Some options contain offloading computation to the cloud, however this introduces points reminiscent of latency and reliance on web connectivity. Different groups are engaged on growing extra environment friendly LLMs tailor-made for robotics, although scaling these options remains to be a technical problem.

As embodied AI turns into extra autonomous, moral issues additionally come up. Who’s accountable if a robotic makes a mistake that causes hurt? How will we guarantee the security of robots working in delicate environments, reminiscent of hospitals? Moreover, the potential for job displacement because of automation is a societal concern that must be addressed by means of considerate insurance policies and oversight.

The Backside Line

Giant language fashions are revitalizing embodied AI, turning robots into machines able to understanding us, reasoning by means of issues, and adapting to sudden conditions. These developments—from pure language processing to multimodal sensing—are making robots extra versatile and accessible. As we see extra real-world deployments, the fusion of LLMs and embodied AI is shifting from a imaginative and prescient to actuality. Nonetheless, challenges reminiscent of accuracy, computational calls for, and moral issues stay, and overcoming these might be key to shaping the way forward for this expertise.

Norway’s New Industrial Coverage Pivots Away From Hydrogen For Vitality



Join day by day information updates from CleanTechnica on e mail. Or comply with us on Google Information!


In 2020, Norway jumped into hydrogen prefer it was the following North Sea oil rush. The federal government launched a nationwide hydrogen technique stuffed with ambitions and buzzwords, envisioning a rustic the place blue and inexperienced hydrogen would decarbonize ships, factories, possibly even houses. In 2021, they doubled down with a hydrogen roadmap that talked about creating home markets and positioning Norway as a serious European provider. They earmarked practically NOK 1 billion—about $100 million USD—for numerous hydrogen pilots and infrastructure investments. It wasn’t pocket change, however it additionally wasn’t remotely near what would have been wanted to make hydrogen power economically viable at scale. Nonetheless, the hydrogen hype cycle was in full spin, and Norway, like a lot of Europe, was alongside for the experience.

The issue, as all the time, was that physics and economics refused to learn the press releases. Hydrogen as an power vector is spectacularly inefficient. Inexperienced hydrogen eats up large volumes of unpolluted electrical energy, with two-thirds of the power disappearing between technology and last use. Blue hydrogen is rather less power-hungry however requires huge infrastructure and comes saddled with upstream methane leaks and a carbon seize system that by no means fairly captures what it guarantees. Norway, a rustic blessed with hydropower and an extended custom of engineering pragmatism, ought to have seen this coming. However the attract of exporting decarbonization in a tank was simply too seductive.

To be scrupulously truthful about Norway’s pure fuel, it is among the greatest engineered and most leak free extraction, processing and distribution programs on this planet, so it has fewer points concerning the full lifecycle greenhouse fuel emissions of blue hydrogen than many different geographies, notably the USA, whose fossil gas trade has the best methane emissions of any on this planet in each absolute and relative phrases by fairly a margin.

By 2024, the shine had worn off. Norway’s new industrial technique, Meld. St. 16, which a Norwegian contact flipped to me right now—thank FSM for web translation—didn’t kill hydrogen outright, however it positively stopped setting a spot for it on the grown-up desk. Useless-end maritime fuels like ammonia nonetheless bought a nod and a few funding from Enova—about $75 million USD throughout 5 initiatives—however the remainder of the hydrogen dream was quietly filed below “too arduous, too costly, too speculative.” The technique speaks in hushed tones about immature markets, poor cost-competitiveness, electrical energy constraints, and transport challenges. That’s bureaucrat-speak for “this isn’t going to work.” The grand visions of hydrogen hubs and European exports have given option to actuality: there’s no demand, no clients, and no monetary logic.

A part of the issue was that Norway’s premise was plentiful electrical energy, however Norway’s energy surplus isn’t what it was once. Demand is hovering because the nation electrifies every little thing from oil platforms to automotive chargers to knowledge facilities. In the meantime, hydropower—the historic spine—has maxed out. They’ve dammed most of what will be dammed, and local weather volatility isn’t serving to reservoir ranges. Wind energy? That stalled after NIMBYs realized 180-meter generators weren’t invisible. Add in export obligations by way of interconnectors to Germany and the UK, and also you’ve bought electrons flowing out simply when home industries are begging for extra. Even the place provide exists, the grid isn’t maintaining—factories are ready years for connections.

A few of Norway’s flagship hydrogen initiatives have been meant to be recreation changers. As a substitute, they turned cautionary tales. The deliberate Aukra blue hydrogen facility, backed by Shell, was scrapped when no consumers materialized. Billions in infrastructure, and no one needed the product. It was speculated to be a cornerstone of Norway’s hydrogen export desires. Seems, nobody in Europe needed to pay a premium for a molecule that loses half its power earlier than it even reaches a pipeline.

Then there was the Hellesylt Hydrogen Hub, pitched as a full-circle inexperienced power ecosystem: electrolysis, hydrogen storage, and zero-emissions ferries all powered by clear hydropower. The buzzwords have been robust, the diagrams slick. However after years of consultant-heavy studies and no critical off-take agreements, it too quietly evaporated. Like many of those initiatives, it collapsed not from opposition however from inaction—dying by silence.

To the notice about consultants, apparently Norway’s hydrogen trade has about 1,100 folks in it and a full 50% are consultants. That gravy practice has ended, so that they’ll need to search for actual jobs now.

I took a have a look at Norway’s maritime hydrogen push in my December 2024 piece Extra Hydrogen Maritime Trials Floor from the Sargasso Sea.” What I discovered wasn’t fairly. Norway, a rustic with a globally revered shipbuilding trade—ranked sixth in Europe primarily based on income—and a few of the most electrified transport wherever, had poured a piece of its innovation capital right into a gas that by no means made sense for its use case.

The MF Hydra was speculated to be the crown jewel—the world’s first liquid hydrogen-powered ferry. However if you pull again the curtain, it’s a logistics nightmare and a local weather head pretend. The hydrogen is liquefied in Germany, trucked greater than 1,300 kilometers to Norway, can’t even undergo tunnels and has to get particular permission to take ferries at particular low-traffic occasions due to security restrictions. The top consequence? The MF Hydra emits about twice as a lot CO₂ full lifecycle because the diesel ferry it was supposed to interchange. Its power prices are roughly ten occasions increased, its emissions are 40 occasions increased and it travels slower than Norway’s well-proven battery-electric ferries.

And this isn’t a one-off glitch—it’s systemic. Norway thought hydrogen would give its shipbuilders a technological edge within the age of decarbonization. As a substitute, they ended up chasing a dead-end gas that’s outperformed by batteries in nearly each ferry route they function. I in contrast it to what British Columbia is doing in Canada: investing straight in battery-electric ferries and increasing shore energy. No unique gas. No costly infrastructure. Simply clear, quiet, quick vessels that work.

Equinor, for its half, learn the room. In late 2024, it scrapped plans to construct the world’s first offshore hydrogen pipeline to Germany. The $3 billion infrastructure mission collapsed below its personal weight, largely as a result of nobody needed the product on the different finish. Across the identical time, Equinor slashed its deliberate investments within the power transition by 50 % and walked again its clear power capability targets. It’s nonetheless claiming that it’s going to be web zero by 2050, however as a substitute of doing it by pivoting to renewables, it’s claiming it’s going to do it by lowering emissions from its fossil gas extraction, processing, refining and distribution and from burying CO2.  The retreat isn’t tactical—it’s a full-blown give up with a well mannered press launch.

Norway hasn’t utterly buried hydrogen, however the grave is dug, the coffin’s half-lowered, and somebody simply forgot the final shovelful of grime. The nationwide technique not positions hydrogen as the way forward for power. It’s a contingency plan at greatest, a backup dancer searching for a stage.

What’s most telling is that Norway isn’t framing this as a failure. It’s doing one thing way more Scandinavian: quietly altering course with out drawing consideration to the truth that the unique plan was flawed. The federal government nonetheless funds some hydrogen tech improvement and nonetheless talks about worth chains and innovation. Nevertheless it has stopped pretending that hydrogen will substitute electrons for heating, transportation, or grid-scale storage. And thank FSM for that.

Hydrogen for power is useless. Not wounded. Not stumbling. Useless. Norway has simply chosen a quiet funeral over a dramatic eulogy. Equinor’s pipeline cancellation, the shift in authorities messaging, and the collapse of early flagship initiatives all level in a single path: a know-how that promised every little thing and delivered nothing. Norway’s engineers, economists, and power planners are lastly aligning with the legal guidelines of physics. The one factor left is to say it out loud.

Whether or not you will have solar energy or not, please full our newest solar energy survey.



Chip in just a few {dollars} a month to assist assist unbiased cleantech protection that helps to speed up the cleantech revolution!


Have a tip for CleanTechnica? Need to promote? Need to counsel a visitor for our CleanTech Speak podcast? Contact us right here.


Join our day by day e-newsletter for 15 new cleantech tales a day. Or join our weekly one if day by day is just too frequent.


Commercial



 


CleanTechnica makes use of affiliate hyperlinks. See our coverage right here.

CleanTechnica’s Remark Coverage




Hackers Utilizing E-Crime Device Atlantis AIO for Credential Stuffing on 140+ Platforms

0


Mar 26, 2025Ravie LakshmananPassword Safety / Cybercrime

Hackers Utilizing E-Crime Device Atlantis AIO for Credential Stuffing on 140+ Platforms

Risk actors are leveraging an e-crime software referred to as Atlantis AIO Multi-Checker to automate credential stuffing assaults, in keeping with findings from Irregular Safety.

Atlantis AIO “has emerged as a robust weapon within the cybercriminal arsenal, enabling attackers to check tens of millions of stolen credentials in fast succession,” the cybersecurity firm mentioned in an evaluation.

Credential stuffing is a kind of cyber assault during which an adversary collects stolen account credentials, usually consisting of lists of usernames or e mail addresses and passwords, after which makes use of them to realize unauthorized entry to person accounts on unrelated programs via large-scale automated login requests.

Cybersecurity

Such credentials may very well be obtained from a knowledge breach of a social media service or be acquired from underground boards the place they’re marketed on the market by different menace actors.

Credential stuffing can also be completely different from brute-force assaults, which revolve round cracking passwords, login credentials, and encryption keys utilizing a trial and error methodology.

Atlantis AIO, per Irregular Safety, gives menace actors the power to launch credential stuffing assaults at scale through pre-configured modules for concentrating on a spread of platforms and cloud-based providers, thereby facilitating fraud, knowledge theft, and account takeovers.

“Atlantis AIO Multi-Checker is a cybercriminal software designed to automate credential stuffing assaults,” it mentioned. “Able to testing stolen credentials at scale, it may rapidly try tens of millions of username and password combos throughout greater than 140 platforms.”

E-Crime Tool Atlantis AIO

The menace actors behind this system additionally declare that it is constructed on “a basis of confirmed success” and that they’ve 1000’s of happy purchasers, whereas assuring clients of the safety ensures baked into the platform in an effort to preserve their buy non-public.

“Each characteristic, replace, and interplay is crafted with meticulous consideration to raise your expertise past expectations,” they state within the official commercial, including “we frequently pioneer options that drive unprecedented outcomes.”

Targets of Atlantis AIO embrace e mail suppliers like Hotmail, Yahoo, AOL, GMX, and Net.de, in addition to e-commerce, streaming providers, VPNs, monetary establishments, and meals supply providers.

Cybersecurity

One other notable side of the software is its potential to conduct brute-force assaults in opposition to the aforementioned e mail platforms and automate account restoration processes related to eBay and Yahoo.

“Credential stuffing instruments like Atlantis AIO present cybercriminals with a direct path to monetizing stolen credentials,” Irregular Safety mentioned.

“As soon as they achieve entry to accounts throughout varied platforms, attackers can exploit them in a number of methods — e.g., promoting login particulars on darkish internet marketplaces, committing fraud, or utilizing compromised accounts to distribute spam and launch phishing campaigns.”

To mitigate the account takeover dangers posed by such assaults, it is advisable to enact strict password guidelines and implement phishing-resistant multi-factor authentication (MFA) mechanisms.

Discovered this text attention-grabbing? Comply with us on Twitter and LinkedIn to learn extra unique content material we put up.



“Crocodilus” A New Malware Concentrating on Android Gadgets for Full Takeover

0


Researchers have uncovered a harmful new cellular banking Trojan dubbed Crocodilus actively concentrating on monetary establishments and cryptocurrency platforms.

The malware employs superior methods like distant machine management, stealthy overlays, and social engineering to steal delicate knowledge, marking a major escalation in cellular menace sophistication.

Early campaigns concentrate on banks in Spain and Turkey, however consultants warn of imminent international growth because the malware evolves.

Crocodilus Debuts With Superior Machine-Takeover Capabilities

Crocodilus distinguishes itself from older banking Trojans like Anatsa or Octo by incorporating “hidden” distant management options from its inception.

As soon as put in by way of a dropper that bypasses Android 13+ safety, the malware abuses Accessibility Providers to observe machine exercise and deploy malicious overlays.

These overlays mimic professional banking apps, tricking customers into coming into credentials, that are harvested in actual time.

A novel “black display screen overlay” conceals fraudulent transactions by masking the machine display screen whereas muting audio, guaranteeing victims stay unaware of unauthorized actions.

Crocodilus additionally makes use of Accessibility Logging a superset of conventional keylogging to seize each textual content change and UI factor displayed, together with one-time passwords (OTPs) from apps like Google Authenticator. This allows attackers to bypass multi-factor authentication seamlessly.

Proof inside Crocodilus’ code factors to Turkish-speaking builders, with debug messages and tags like “sybupdate” suggesting potential hyperlinks to “sybra”—a menace actor beforehand linked to Ermac, Hook, and Octo malware variants.

Nevertheless, researchers warning that “sybra” could possibly be a buyer testing Crocodilus moderately than its creator, highlighting the malware’s possible availability in underground markets.

The Trojan’s infrastructure already helps dynamic concentrating on, permitting operators to push up to date overlay templates and app goal lists by way of its C2 server.

Early targets embody main Spanish banks, Turkish monetary apps, and cryptocurrency wallets like Bitcoin Pockets and Belief Pockets.

ThreatFabric anticipates speedy diversification of targets as Crocodilus beneficial properties traction amongst cybercriminals.

Social Engineering Lures Victims into Surrendering Crypto Keys

In a devious twist, Crocodilus manipulates cryptocurrency customers into voluntarily revealing pockets restoration phrases.

After stealing a pockets’s PIN by way of an overlay, the malware shows a faux warning: “Again up your pockets key within the settings inside 12 hours. In any other case, the app will likely be reset…”

Panicked victims then navigate to their seed phrase, which Accessibility Logger captures and transmits to attackers, which grants full management over wallets, enabling instantaneous asset theft.

In line with the Report, Crocodilus’ speedy maturation underscores the inadequacy of conventional antivirus instruments in opposition to fashionable banking Trojans.

ThreatFabric urges monetary establishments to undertake behavior-based detection and machine threat profiling to establish compromised gadgets.

Customers are suggested to keep away from sideloading apps, scrutinize app permissions, and mistrust pressing safety warnings with out verification.

As cellular threats develop extra refined, the battle in opposition to fraud more and more hinges on disrupting the social engineering ways that make instruments like Crocodilus devastatingly efficient.

Discover this Information Attention-grabbing! Comply with us on Google InformationLinkedIn, and X to Get Immediate Updates!