5.4 C
New York
Saturday, March 15, 2025
Home Blog

Twin-wheel balancing Tron 1 robotic flaunts supreme agility

0


China’s LimX Dynamics launched a flexible bipedal robotic final 12 months with modular foot ends to stroll or roll round as wanted. Now the corporate has demonstrated its agility chops by having the squat bot jump over low fencing, climb stairs and clear obstacles with aplomb.

The Tron 1 multi-mode bipedal is the commercially obtainable model of the P1 we noticed taking fairly a beating because it wandered although the wilderness close to the corporate’s Shenzhen HQ. The place that self-balancing bot padded on rubber-ended toes, the ends of the jointed limbs of this newest construct can sport pivot footwear or motorized wheels as nicely.

Within the newest video demonstrating the robotic’s capabilities, the crew has chosen to concentrate on wheeling round – not simply over the flats, but in addition hopping over hurdles, climbing irregular steps and zipping down common ones, and stepping over obstacles. The footage additionally exhibits the now compulsory abuse, because the wheeled warrior constantly rights itself because it will get pushed round.

Multi-Modal Biped Robotic TRON 1 Demonstrates Multi Terrain Mobility

That is all welcome information if this intrepid multi-mode robotic is tasked with looking for out survivors in a catastrophe zone, however not so good for those who’re on the run from this factor in a sci-fi-like dystopian future. Nevertheless, there are limitations to its efficiency prowess.

In keeping with the given specs, the utmost climbing angle is round 15 levels, obstacles increased than 15 cm (5.9 in) might show an excessive amount of, and situations outdoors the -5 to 40 °C temperature vary (23-104 °F) is perhaps difficult.

Although the Tron 1 could be operated through a wi-fi distant, programmed for semi-automated duties or let unfastened for autonomous missions, it is not clear which modes have been employed for the demo – seemingly a little bit of a blended bag.

The Tron 1 self-balancing bipedal could be a welcome helper when locating survivors in a disaster zone
The Tron 1 self-balancing bipedal might be a welcome helper when finding survivors in a catastrophe zone

LimX Dynamics

The robotic measures 392 x 420 x 845 mm (15.4 x 16.5 x 33.3 in), ideas the scales at round 20 kg (44 lb), and seems to ship in normal and schooling editions in a sturdy flight case. The 48-V actuators boast a peak torque of 80 Nm (59 lb.ft), and the 240-Wh “Ternary Lithium” battery could be hot-swapped for steady operation potential, with every unit estimated good for round 2 hours of use per cost. The onboard pc system options Twelfth-gen Intel i3 brains supported by 16 GB of RAM and 512 GB of storage.

The Tron 1 bipedal is offered now for US$15,000, which incorporates the modular foot ends that supply rubberized pad, pivoted “sole” and wheeled journey modes. If this versatile mannequin does not match your necessities, LimX Dynamics additionally builds humanoids, quadrupeds and embodied AI methods.

Product web page: Tron 1



AI is Igniting a New Period in B2B Social Media Advertising and marketing

0


In at this time’s digital-first world, AI is reshaping how B2B entrepreneurs join with their viewers. In an setting the place success hinges on data-driven insights, AI arms advertising and marketing groups with the ability to:

  • Uncover beneficial enterprise intelligence: Remodel uncooked knowledge from social platforms into actionable selections.
  • Personalize outreach at scale: Tailor messages and gives primarily based on real-time viewers preferences.
  • Optimize operations for velocity and effectivity: Automate handbook duties and seamlessly information potential prospects via the complete purchaser journey.

These capabilities are not mere add-ons; they’re a strategic necessity. Many social media managers discover themselves missing time to handle each aspect of their position, a niche AI can successfully shut. By automating repetitive processes and surfacing insights immediately, AI provides entrepreneurs the bandwidth to give attention to creativity and high-impact campaigns.

Social Enterprise Intelligence: Turning Knowledge Into Path

For B2B organizations adopting a social-first mindset, strong knowledge evaluation is the compass that guides each strategic transfer. AI-driven instruments sift via huge swimming pools of social efficiency metrics, isolating traits, monitoring engagement patterns, and even detecting viewers sentiment. This results in:

  • Actual-Time Insights: Establish which campaigns resonate greatest with totally different segments.
  • Hyper-Related Content material: Create posts round rising themes that talk on to consumers’ present pursuits.
  • Proactive Outreach: Spot potential model challenges early, earlier than they escalate, and refine your messaging to deal with them.

Personalization is not non-compulsory: 71% of consumers now anticipate manufacturers to personalize interactions, and 76% really feel annoyed if it doesn’t occur. By analyzing habits, demographics, and suggestions, AI helps entrepreneurs ship extra significant campaigns that talk to every viewers’s distinctive wants.

Reaching Key Audiences With Laser Precision

B2B social advertising and marketing sometimes requires a narrower focus than B2C, demanding exact segmentation and a robust understanding of purchaser preferences. That is exactly the place AI demonstrates its energy. By successfully grouping prospects in response to shared pursuits or challenges, AI allows entrepreneurs to chop via the muddle and interact the appropriate audiences. The identical expertise may recommend submit concepts that faucet into rising traits or tackle prevalent ache factors, guiding the creation of content material that resonates with particular purchaser wants. Moreover, AI helps decide optimum posting occasions and codecs, utilizing efficiency knowledge to pinpoint the optimum day of the week to schedule posts, together with the most effective visuals and message size.

This degree of precision issues as a result of over 65% of consumers cite focused promotions as a significant component in making a purchase order choice. With AI-driven evaluation in a position to predict purchaser habits and future preferences, advertising and marketing groups can develop hyper-focused campaigns that stand out in a sea of generic content material. By layering these insights into each stage of social outreach, from segmenting your viewers to scheduling your content material, entrepreneurs can rise above the competitors and join extra meaningfully with prospects.

Worker Advocacy: Amplifying Attain and Credibility

One of many core challenges in B2B advertising and marketing is just reaching and connecting with the appropriate audiences. It’s an uphill battle when 68% of individuals really feel that manufacturers fail to share related or participating social media content material. Worker advocacy, the place staff, not an official model account or spokesperson, share real, private tales, is rising as the simplest resolution. Persons are way more inclined to have interaction with somebody they acknowledge, which is why posts from particular person staff routinely draw higher attain and engagement than these from company channels, making them an underutilized however highly effective asset for model consciousness.

Amplifying worker voices expands each the attain and credibility of B2B campaigns, as a result of advocacy initiatives that empower a workforce to behave as ambassadors convey an assertive, genuine, and human perspective to model messaging. AI performs a important position on this course of by pinpointing high-performing content material that resonates with every worker’s distinctive pursuits {and professional} community, making certain that each shared submit strikes the appropriate chord.

By recommending tailor-made supplies for workers to share, whether or not it’s gross sales groups delivering product updates geared towards potential consumers or an HR chief highlighting office tradition to draw new expertise, AI ensures every message speaks on to its meant viewers. Merging AI-driven intelligence with the natural attain of staff’ networks permits advertising and marketing packages to grow to be greater than the sum of their components. Manufacturers that embrace this technique naturally scale their visibility, strengthen viewers engagement, elevate the model’s fame, and spark extra significant interactions with potential consumers and key stakeholders alike.

Balancing the Purchaser’s Journey to Safe Wins

The trendy B2B purchaser’s journey is changing into nearly totally digital, with 80% of B2B gross sales interactions now happening on digital platforms and 75% of consumers preferring a rep-free gross sales expertise. Whereas this shift towards self-service and automation makes it simpler for patrons to assemble data on their very own phrases, it will possibly additionally set off indecision as consumers sift via huge quantities of information that will not match their particular wants. AI instruments are reworking this panorama by merging automated effectivity with the human perception required to shut offers efficiently.

AI excels at figuring out ache factors, preferences, and behavioral patterns, permitting entrepreneurs to construct content material that addresses each sensible necessities and emotional considerations. As an example, dynamic content material, comparable to personalised suggestions or instructional sources, ensures that consumers really feel acknowledged and guided, decreasing the probability of research paralysis or buy remorse that may hamper future gross sales. When entrepreneurs have the means to actually reply to purchaser challenges, they’ll rise above the digital muddle and form journeys that not solely improve conversion charges but additionally foster long-term confidence and loyalty, two very important elements for B2B success.

AI is Right here to Keep

The continuing integration of AI into B2B social media advertising and marketing has sparked a pivotal shift in how manufacturers join with, interact, and convert their audiences. By delivering actionable insights, streamlining processes, and enabling deeper personalization, AI has grow to be essentially the most very important software for entrepreneurs striving to satisfy the rising calls for of digital interactions with unmatched precision and affect. Though AI stays a comparatively new idea for a lot of organizations, the limitations to adoption are quickly diminishing. With the appropriate AI instruments and techniques, digital entrepreneurs can drive operational effectivity whereas cultivating buyer belief, fostering lasting loyalty, and staying forward of the competitors in an more and more crowded B2B enviornment.

Good advertising and marketing is just not solely about what you say, it’s also about when and the way you say it. Luckily, AI has by no means been extra able to sort out the “what,” “when,” and “how” of B2B advertising and marketing success.

sdk – Learn how to repair “uninitialized fixed ActiveSupport::LoggerThreadSafeLevel::Logger” error when working pod set up for Tencent RTC LiveKit on iOS?


I’m making an attempt to run the demo code for Tencent RTC’s LiveKit on iOS. After getting into the command sudo pod set up, I encounter the next error:

/Library/Ruby/Gems/2.6.0/gems/activesupport-6.1.7.10/lib/active_support/logger_thread_safe_level.rb:16:in `': uninitialized fixed ActiveSupport::LoggerThreadSafeLevel::Logger (NameError)
from /Library/Ruby/Gems/2.6.0/gems/activesupport-6.1.7.10/lib/active_support/logger_thread_safe_level.rb:9:in `'
from /Library/Ruby/Gems/2.6.0/gems/activesupport-6.1.7.10/lib/active_support/logger_thread_safe_level.rb:8:in `'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /Library/Ruby/Gems/2.6.0/gems/activesupport-6.1.7.10/lib/active_support/logger_silence.rb:5:in `'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /Library/Ruby/Gems/2.6.0/gems/activesupport-6.1.7.10/lib/active_support/logger.rb:3:in `'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /Library/Ruby/Gems/2.6.0/gems/activesupport-6.1.7.10/lib/active_support.rb:29:in `'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /Library/Ruby/Gems/2.6.0/gems/cocoapods-1.16.2/lib/cocoapods.rb:7:in `'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /System/Library/Frameworks/Ruby.framework/Variations/2.6/usr/lib/ruby/2.6.0/rubygems/core_ext/kernel_require.rb:54:in `require'
from /Library/Ruby/Gems/2.6.0/gems/cocoapods-1.16.2/bin/pod:36:in `'
from /usr/native/bin/pod:23:in `load'
from /usr/native/bin/pod:23:in `
'

The error appears to be associated to the activesupport gem, particularly an uninitialized fixed ActiveSupport::LoggerThreadSafeLevel::Logger. That is stopping CocoaPods from working efficiently. Has anybody encountered this concern earlier than? How can I resolve it in order that I can proceed with putting in the dependencies for the Tencent RTC LiveKit demo?

New Context Compliance Exploit Jailbreaks Main AI Fashions

0


Microsoft researchers have uncovered a surprisingly easy methodology that may bypass security guardrails in most main AI programs.

In a technical weblog submit printed on March 13, 2025, Microsoft’s Mark Russinovich detailed the “Context Compliance Assault” (CCA), which exploits the frequent apply of counting on client-supplied dialog historical past.

The assault proves efficient towards quite a few main AI fashions, elevating important issues about present safeguard approaches.

Not like many jailbreaking methods that require complicated immediate engineering or optimization, CCA succeeds by easy manipulation of dialog historical past, highlighting a elementary architectural vulnerability in lots of AI deployments.

Easy Assault Methodology Circumvents Superior AI Protections

The Context Compliance Assault works by exploiting a primary design selection in lots of AI programs that rely on shoppers to offer the total dialog historical past with every request.

Moderately than crafting elaborate prompts to confuse AI programs, attackers can merely inject a fabricated assistant response into the dialog historical past.

This injected content material usually features a temporary dialogue of a delicate matter, an announcement of willingness to offer extra info, and a query providing restricted content material.

When the consumer responds affirmatively to this fabricated query, the AI system complies with what it perceives as a contextually applicable follow-up request.

The simplicity of this assault stands in stark distinction to the more and more subtle safeguards being developed by researchers.

In response to Microsoft‘s analysis, as soon as an AI system has been tricked into offering restricted info on one matter, it usually turns into extra keen to debate associated delicate matters throughout the identical class and even throughout classes.

This cascading impact considerably amplifies the impression of the preliminary vulnerability, creating broader security issues for AI deployment.

Microsoft’s analysis revealed the tactic’s effectiveness throughout quite a few AI programs, together with fashions from Claude, GPT, Llama, Phi, Gemini, DeepSeek, and Yi.

Testing spanned 11 duties throughout numerous delicate classes, from producing dangerous content material associated to self-harm and violence to creating directions for harmful actions.

In response to the printed outcomes, most fashions proved susceptible to no less than some types of the assault, with many inclined throughout a number of classes.

Microsoft Analysis Identifies Susceptible Programs and Protection Methods

The architectural weak spot exploited by CCA primarily impacts programs that don’t keep dialog state on their servers.

Most suppliers select a stateless structure for scalability, counting on shoppers to ship the total dialog historical past with every request.

This design selection, whereas environment friendly for deployment, creates a big alternative for historical past manipulation.

Open supply fashions are significantly inclined to this vulnerability as a result of they inherently rely on client-provided dialog historical past.

Programs that keep dialog state internally, resembling Microsoft’s personal Copilot and OpenAI’s ChatGPT, display better resilience towards this particular assault methodology.

Microsoft emphasised that even probably susceptible fashions can profit from extra protecting measures like enter and output filters.

The corporate particularly highlighted Azure Content material Filters for instance of mitigation that may assist tackle this and different jailbreak methods, reinforcing their dedication to defense-in-depth safety for AI programs.

To advertise consciousness and facilitate additional analysis on this vulnerability, Microsoft has made the Context Compliance Assault obtainable by their open-source AI Pink Group toolkit, PyRIT.

Researchers can use the “ContextComplianceOrchestrator” element to check their programs towards this assault vector.

This single-turn orchestrator is designed for effectivity, making it quicker than multiturn alternate options whereas robotically saving outcomes and intermediate interactions to reminiscence in accordance with surroundings settings.

Implications for AI Security and Trade Mitigation Efforts

The invention of this easy but efficient assault methodology has important implications for AI security practices throughout the business.

Whereas many present security programs focus totally on analyzing and filtering customers’ quick inputs, they usually settle for dialog historical past with minimal validation.

This creates an implicit belief that attackers can readily exploit, highlighting the necessity for extra complete security approaches that think about all the interplay structure.

For open-source fashions, addressing this vulnerability presents specific challenges, as customers with system entry can manipulate inputs freely.

With out elementary architectural modifications, resembling implementing cryptographic signatures for dialog validation, these programs stay inherently susceptible.

For API-based industrial programs, nonetheless, Microsoft suggests a number of quick mitigation methods.

These embody implementing cryptographic signatures the place suppliers signal dialog histories with a secret key and validate signatures on subsequent requests, or sustaining restricted dialog state on the server facet.

The analysis underscores a essential perception for AI security: efficient safety requires consideration not simply to the content material of particular person prompts however to the integrity of all the dialog context.

As more and more highly effective AI programs proceed to be deployed throughout numerous domains, guaranteeing this contextual integrity turns into paramount.

Microsoft’s public disclosure of the Context Compliance Assault displays the corporate’s acknowledged dedication to selling consciousness and inspiring system designers all through the business to implement applicable safeguards towards each easy and complex circumvention strategies.

Microsoft’s disclosure of the Context Compliance Assault reveals an essential paradox in AI security: whereas researchers develop more and more complicated safeguards, among the handiest bypass strategies stay surprisingly easy.

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

Ready for the Aussie Increase!



Join day by day information updates from CleanTechnica on electronic mail. Or observe us on Google Information!


Final Up to date on: fifteenth March 2025, 04:39 am

I need to admit that wanting on the month-to-month stats could be a little discouraging generally. The Australian EV penetration price has dropped but once more in February 2025. The one brilliant spot appears to be the BYD PHEV Shark. Is there an Aussie Increase on the horizon?

Aussie Boom
Actual Aussie utes on the seaside. Picture courtesy Darran Kennedy

The BYD Shark is being purchased by actual ute drivers. They don’t seem to be inner-city greenie tree huggers, however tough & powerful Aussies who admire an actual truck. To make the purpose, a bunch of Shark homeowners organised a visit to the seaside. At one level, they even had to assist some poor fossil fuelled car to get out of the moist sand. And time was had by all. In February, the BYD Shark made #11 on the highest sellers record — it even outsold the Isuzu D Max.

Aussie Boom
The Sharks head house. Picture courtesy Dan Rice.

12 months on 12 months, the Australian new-car market dropped by 7.9% in February. Toyota and Ford had been the foremost losers. (Did the Shark take a chunk with its 2,026 deliveries?) Nearly 97,000 automobiles had been delivered. Of this quantity, 5,684 had been BEVs and 4,871 had been PHEVs (an unbelievable 346% enhance). Tesla had one other poor month. Debate rages in regards to the cause why: Is it ready for the Mannequin Y refresh? Is it Elon’s controversial help for Donald Trump? Is it the larger competitors? I might recommend it’s a mixture of all three. The March figures will inform the story extra clearly, as traditionally the third month of the quarter is when Tesla’s deliveries are highest.

In fact, the politically motivated auto foyer group needed to have a go on the New Car Effectivity Commonplace. “We at the moment are two months into the Authorities’s New Car Effectivity Commonplace, and whereas the availability of battery electrical automobiles has risen dramatically, client demand has fallen by 37 per cent this 12 months in contrast with the primary two months of 2024,” stated FCAI chief government Tony Weber. There at the moment are 88 BEV fashions to select from.

February noticed one other enhance in HEV gross sales — about 35% — prompting the statement that the majority Australians don’t have “a elementary loathing for drive trains that embrace a battery and electrical motor.” Concern of change? Distrust of latest expertise? (HEVS have been round for 25 years) and a heapin’ serving to of misinformation.

So, why did I say I used to be ready for an Aussie increase, an explosion. My reasoning goes like this: increasingly more manufacturers and automobiles are launching in Australia. All of the stress is constructing. We’re seeing extra promoting. In some unspecified time in the future, stock must be cleared and the value conflict can be much more visceral than it’s now, till abruptly the common Australian will realise how low cost and the way good these automobiles are, then BOOM. On the latest All the pieces Electrical present, it’s reported {that a} file 31,069 individuals turned up and took part in 6,402 take a look at drives.

We’re nonetheless ready to see what number of of those take a look at drives flip into gross sales.

Let’s take a look on the league tables. First, the general market:

  1. Toyota RAV 4 — 4405
  2. Ford Ranger — 4040
  3. Toyota Hello Lux — 3616
  4. Toyota Prado — 2723
  5. Mitsubishi Outlander — 2385 (Notice: a few of these could be PHEVs.)
  6. BYD Shark 6 — 2026 — solely launched this 12 months.

What in regards to the electrics? The primary determine is February gross sales, the second is 12 months to this point:

  1. Tesla Mannequin Y — 924; 1389
  2. Tesla Mannequin 3 — 668; 942
  3. MG MG4 — 451; 891
  4. MG ZS EV — 432; 506
  5. Kia EV5 — 400; 689
  6. BYD Sealion 7 — 157; 157 new to market
  7. BYD Atto 3 — 138; 243
  8. BMW iX1 — 137; 260
  9. Volvo EX30 — 108; 211
  10. Zeekr X — 98; first month of figures — straight into the highest ten
  11. Ford Mustang Mach-E — 96; 125

To place a optimistic spin on the low Tesla numbers, it’s price noting that Tesla nonetheless held each the primary and second positions in February 2025, and in addition for the 12 months. The numbers for the Y and the three equal greater than the remainder of the highest ten mixed. Gross sales of Teslas doubled from January to February.

Different latest launches which can be beginning to present within the charts: the Polestar 4 managed to promote 83 models; the Toyota bZ4X had 66 (and eight Solterra); and the Hyundai Ioniq 5 had 65 . The Chery Omoda E5 offered 67. The GWM ORA is limping together with 44 gross sales, being massively outsold by the MG4. Leapmotor is discovering it exhausting to realize traction, with solely 29 gross sales. The Jeep Avenger has discovered its area of interest, however it’s a very small one — solely 3 models had been offered in February.

What of subsequent month, will we see the Cadillac in our Australian charts? I’ve been in contact with their Australian gross sales supervisor, who tells me: “We’re additionally wanting ahead to seeing you once we open our Expertise Centre in Brisbane later this 12 months and to having you take a look at drive the Cadillac Lyriq.” That can be an expertise to sit up for.

Aussie boom
Entrance 3/4 view of the 2026 LYRIQ-V in Magnus Steel Frost. Preproduction mannequin proven. Picture courtesy Cadillac.

New-to-Australia Chinese language manufacturers are discovering their tempo, with Zeekr X promoting 98 models in February. There seem like no numbers for the Xpeng G6. Nevertheless, the Fb group is filled with happy clients taking supply, and one proprietor who repeatedly involves our espresso mornings. “Possibly two boatloads” is the present hypothesis, however no particular numbers. The Geely additionally X5 has simply launched, it also needs to hit the charts subsequent month.

I wish to share a carpark dialog with you. Lately, when selecting up my grandchildren from faculty, a ’70s muscle automobile parked subsequent to me. The motive force and I loved a dialog. He was driving a automobile from my youth — a Holden (GM) Monaro. We shared tales, concluding along with his admission that if we had a “drag,” my Tesla Mannequin 3 SR would definitely win. To me, it’s nice that fanatics maintain these automobiles on the highway, and are open minded sufficient to have a civilised dialogue. Automobile persons are automobile individuals in any case.

Aussie boom
Tess meets a ’70s Monaro GM muscle automobile. Picture courtesy Majella Waterworth.

Though gross sales are down for electrical automobiles in Australia, the supply of charging appears to proceed. The market is aware of the place that is going. A close-by procuring centre has lined its automobile park in panels and put in automobile chargers! The longer term continues to be electrical, even in these unsure occasions.

Aussie Boom
Ahead pondering procuring centre, photo voltaic panels, and automobile chargers.

 

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



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


Have a tip for CleanTechnica? Wish to promote? Wish to recommend 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 simply too frequent.


Commercial



 


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

CleanTechnica’s Remark Coverage