Scientists have developed a new type of quantum building block, called an erbium molecular qubit, that could allow quantum devices to connect directly to existing fiber-optic networks. In simple terms, these qubits act as a bridge between quantum computers and the internet we already use, potentially letting quantum information travel across the same networks that carry emails and videos today.

The key breakthrough is that these qubits respond to light in the same wavelength range used by telecom networks. That means quantum data can be sent and received using current fiber-optic cables without the need to build entirely new infrastructure.

How It Works

Think of these qubits as tiny switches that can be controlled in two ways: with magnetic fields and with light. This dual control lets scientists read and manipulate quantum information precisely.

Leah Weiss, a researcher at the University of Chicago, explained that these molecules act like a bridge between the magnetic world and the optical world. Using special tools, researchers can control these qubits with extremely fine precision. This makes them useful for connecting quantum processors or sensors to photonic systems—basically, linking quantum devices to the light signals that travel through fiber cables.

Why It’s Important

Current quantum experiments mostly happen in labs, using specialized setups. To make quantum computing and communication practical, we need ways to connect these devices over real-world networks.

Because these erbium qubits work with standard fiber-optic light, they could plug quantum technology directly into the internet we already have. That opens the door to more practical quantum communication, hybrid systems that combine different types of quantum control, and devices that can sense or transmit information in new ways.

David Awschalom, a leading researcher on the project, said this step moves quantum networks closer to reality, making it possible to expand them gradually without rebuilding everything from scratch.

What’s Next

While the results are promising, there’s still work to do. Researchers need to make sure the qubits perform reliably in larger systems, integrate smoothly with computer controllers, and function consistently outside the lab.

Even so, this research hints at a future where quantum technology could work alongside the networks we use every day, bringing quantum computing and communication closer to practical applications.

Peer Review Under Strain

The volume of submissions to major conferences has increased faster than the availability of qualified reviewers. Tools like Citation Check help highlight suspicious references but rely on human verification for final judgment.

The integrity of citations is critical. Misleading or fabricated references can distort the scientific record and influence subsequent research, policy, or technology adoption. Identifying and questioning these discrepancies is part of maintaining a reliable scholarly ecosystem.

Patterns in Hallucinations

Among the 50 flagged papers, patterns included:

Incorrect author lists paired with real paper titles.

Altered journal names, years, and page numbers.

Papers combining multiple sources into a single fabricated citation.

These inconsistencies illustrate how small deviations can propagate unnoticed through peer review if attention to detail is lacking.

What This Means for Research

Citations serve as the connective tissue of scientific work. When references are unreliable, readers lose context, the credibility of research is weakened, and the mechanisms of verification break down. Identifying hallucinations is not just about catching errors, it is about understanding where scholarly practices may need adjustment.

Why Germanium is Making a Comeback

Silicon has been the backbone of semiconductor devices for decades, but as components shrink, they generate more heat and approach performance limits. Germanium, a material used in the first transistors in the 1950s, offers superior electrical properties while remaining compatible with existing silicon manufacturing techniques.

Dr. Maksym Myronov, Associate Professor at the University of Warwick, explained, “Traditional high-mobility semiconductors like gallium arsenide are expensive and hard to integrate. Our compressively strained germanium-on-silicon (cs-GoS) quantum material combines record mobility with industrial scalability, a key step for practical quantum and classical circuits.”

Engineering the Record Mobility

The researchers grew a thin germanium layer on a silicon wafer and applied precise compressive strain. This process produced an exceptionally pure and orderly crystal structure that allows electrical charge to flow with minimal resistance.

Tests revealed a hole mobility of 7.15 million cm²/V·s, compared to roughly 450 cm²/V·s in standard industrial silicon. This massive improvement could enable chips that operate faster, consume less power, and run cooler—critical for AI accelerators, data center servers, and next-gen electronics.

Impact on Quantum and Next-Gen Devices

Dr. Sergei Studenikin from the National Research Council of Canada highlighted that this sets a new benchmark for charge transport in group-IV semiconductors. The discovery is relevant for:

Quantum information systems and spin qubits

Cryogenic controllers for quantum processors

Energy-efficient AI accelerators

Large-scale integrated circuits compatible with silicon

The development strengthens the UK’s leadership in advanced semiconductor materials research and positions germanium-on-silicon technology as a practical route for future high-performance devices.

Journal Reference: Maksym Myronov, Alex Bogan, Sergei Studenikin. Hole mobility in compressively strained germanium on silicon exceeds 7 × 10⁶ cm²V⁻¹s⁻¹. Materials Today, 2025; 90: 314.

Link to Original Release

After a two-year pause, Coinbase India crypto exchange has reopened its platform for registrations. Users can currently trade crypto-to-crypto pairs, but the company announced plans to introduce a fiat on-ramp in 2026. This will allow Indian users to deposit local currency and purchase cryptocurrencies directly through the app.

Coinbase first entered India in 2022 but had to halt support for the Unified Payments Interface (UPI) after the National Payments Corporation of India (NPCI) refused to recognize its operations. In 2023, Coinbase fully ceased operations in the country, asking users to offload accounts.

John O’Loghlen, Coinbase’s APAC director, told India Blockchain Week that the company engaged with the Financial Intelligence Unit (FIU) and registered earlier this year. Early access onboarding started in October, and the app is now fully open for Indian users.

Strategic Move Amid Regulatory Challenges

India’s strict crypto regulations and taxation policies have historically made expansion difficult. The country levies a 30% tax on crypto income without loss offsets and imposes a 1% deduction on each transaction, which can discourage frequent trading.

O’Loghlen said Coinbase hopes future regulatory adjustments will make holding digital assets easier. Despite challenges, Coinbase’s venture arm recently invested in local exchange CoinDCX, valuing it at $2.45 billion post-money.

User Experience and Trust Focus

Coinbase is positioning itself as a secure and trusted exchange in India. O’Loghlen emphasized, “We want users to onboard in minutes, with the same trusted experience they expect from top Indian super apps like Zepto or Flipkart.” The company aims to offer a simple, reliable interface while safeguarding funds.

Team Expansion and Local Presence

Coinbase plans to grow its Indian team beyond 500 employees, hiring for roles in local operations and global market support. The company will leverage its India base to enhance both regional and international crypto initiatives.

Related Laterstack Blockchain Stories

Bitcoin Miner Bitfarm Shifts from Crypto to AI Data Centers

Bitcoin Plunge Shakes Markets, Signals New Link Between Crypto and Stocks

For inquiries, tips, or submissions: laterstack@proton.me

Yoodli, a Seattle-based startup that provides an AI communication platform for professionals, has reached a valuation of over $300 million following a $40 million Series B round led by WestBridge Capital, with participation from Neotribe and Madrona. This follows a $13.7 million Series A round earlier this year, bringing total funding to nearly $60 million.

The company focuses on helping users improve public speaking, interview skills, sales pitches, and leadership coaching. Yoodli uses AI to run realistic role-play simulations, giving structured and repeatable practice without replacing human coaches.

Varun Puri, co-founder of Yoodli along with former Apple engineer Esha Joshi in 2021, previously worked at Google’s X division. He says the platform addresses communication challenges, especially for international students and young professionals adjusting to new work environments.

Enterprise Growth Drives Adoption

Yoodli initially launched as a consumer-facing public speaking tool but quickly pivoted to enterprise adoption. Companies including Google, Snowflake, Databricks, RingCentral, and Sandler Sales now rely on the platform for employee and partner training. Coaching firms such as Franklin Covey and LHH integrate Yoodli with their own methodologies.

The platform works with multiple large language models, including Google Gemini and OpenAI GPT, and supports embedding into enterprise software or direct browser access. Yoodli also supports multiple languages, including Korean, Japanese, French, Canadian French, and several Indian languages.

Human Guidance Remains Central

Puri emphasized that the platform enhances rather than replaces human coaching. AI can help users reach advanced levels, but human feedback remains essential for authenticity and growth. Yoodli does not offer a mobile app to avoid disrupting the learning experience and keep training streamlined.

Funding and Leadership Updates

Between Series A and B, Yoodli saw a 50 percent increase in the number of role-plays run and total user engagement time. Recurring revenue grew by 900 percent over 12 months, though exact figures were not disclosed.

Recent hires include Josh Vitello as CRO, Andy Larson as CFO, and Padmashree Koneti as CPO. Strong performance metrics, enterprise adoption, and senior hires contributed to the successful Series B round.

Scaling Globally

The funding will expand AI coaching, analytics, and personalization tools, particularly in enterprise learning and professional development. Yoodli plans to hire across product, AI research, and customer success teams while expanding into the Asia-Pacific market and deepening its U.S. presence.

The startup has roughly 40 employees and continues to focus on vertical-specific training and customization for enterprise clients.

Aptera Solar Electric Vehicle Moves Into Validation Builds After IPO

Aptera released its first post-IPO progress update this week, confirming that the Aptera solar electric vehicle has entered the validation build stage inside its Southern California facility. After years of delays and early prototypes, the company is finally producing limited pre-production vehicles intended for testing, durability studies and process refinement.

For longtime watchers of the company, this moment matters less for what the vehicles are and more for the fact that they now exist as production units rather than concept visuals. Validation builds are the first checkpoint where a hardware startup must prove repeatability, even at a small scale.

Aptera’s founders say these units allow the team to stress test the composite shell, solar panel integration and in-wheel motors. They also shared a look at the updated interior and assembly tooling, giving the most concrete snapshot of progress since the company’s public listing.

A Different Approach in an EV Market That Keeps Getting Heavier

The Aptera solar electric vehicle has always positioned itself as an efficiency-first alternative in a market trending toward larger batteries, larger screens and increasingly complex software stacks. The company still believes the fastest way to reduce charging demand is not by building more infrastructure but by designing vehicles that need less of it in the first place.

This philosophy remains unusual in mainstream automaking, where efficiency gains often take a back seat to feature expansion. Aptera’s design challenges that assumption and pushes people to look at energy use from the input side instead of the output side. The company’s supporters see this as a necessary correction. Critics still see it as an experiment.

But with charging queues growing in major cities and material costs rising across the EV supply chain, the broader landscape is starting to tilt toward the kind of efficiency Aptera specializes in.

Public Status Means Less Room for Grand Promises

Going public introduces a new set of expectations. Progress updates must reflect actual production work, not projections. Aptera’s quieter, more measured tone this year reflects that shift. The company describes its current stage as a controlled scale-up, not a launch, and says there is no rush to oversell timelines.

The update shows a more grounded view of the challenges ahead. The team still needs to stabilize composite manufacturing, validate its solar components with suppliers and maintain predictable cash flow during early ramp. For any EV startup, these steps tend to be more difficult than the engineering itself.

A Market That Finally Makes Efficiency Practical

The EV sector in 2025 looks different from the one Aptera entered over a decade ago. High grid demand, rising battery costs and recurring software issues across the industry have pushed consumers to reconsider what an electric vehicle should be optimized for.

This shift gives Aptera a more realistic pathway than it had in previous cycles. The company is not trying to compete directly with family SUVs or high-performance EVs. It is building an ultra-efficient commuter vehicle that reduces energy consumption at the design level. As the company puts it, the best way to avoid charging headaches is to reduce the need for charging in the first place.

The Real Test Comes Next

The success of the Aptera solar electric vehicle now depends on whether the company can transition from low-volume validation builds to repeatable manufacturing. This is the stage where many hardware startups hit structural friction. Validation vehicles can be built by hand. Production vehicles must be built by process.

Aptera says the next several months will focus on assembly consistency, supplier alignment and continued testing of solar coverage and thermal performance. The company also plans to release more frequent updates now that it is publicly traded.

Whether Aptera reaches full production or not, the vehicle forces a useful question into the broader EV conversation. Instead of asking how much power an EV can store, it asks how much power an EV actually needs. That alone puts this startup in an interesting position as the industry enters its next phase.

A School System That Went All In on Devices

The Los Angeles Unified School District provides iPads and Chromebooks to students beginning as early as kindergarten. Some schools require students to take those devices home every day. Others tie key assignments to iPad based platforms. Parents who do not want their children on certain apps can opt out of YouTube and some Google services, but they cannot always opt out of the assignments themselves.

What began as a digital equity effort in the 2010s grew into a full systemwide commitment during the COVID era. Screens became the default instructional tool. But now families are watching what happens when students are overwhelmed, overstimulated, or simply more entertained by games than lessons.

During district listening sessions, nearly every parent who spoke criticized the amount of time children spend on screens in class. Many described changes in behavior. Others pointed to falling grades or isolation. Several parents asked why kindergarteners were handed device contracts warning them not to meet people from the internet. The officials running the sessions declined to answer.

The Return of an Old Scandal

District veterans have not forgotten the troubled 2014 initiative that attempted to give every student a tablet. That project collapsed after concerns about the bidding process and the ease with which students bypassed security protocols. The district now leaves device management to each school, but the result is wide inconsistency.

Some students spend half their day staring at screens. Others spend much less. Many classrooms rely heavily on software like iReady, which adapts lessons using proprietary algorithms that teachers and parents cannot review. Teachers describe it as a black box that makes it difficult to understand what children are being asked or how to support them.

Teachers Are Feeling the Strain Too

Some educators report that mandatory software requirements conflict with what they actually teach. Armaghan Khan, a science teacher, says his students often use answers generated by chatbots and circumvent the school’s monitoring tool by creating new user profiles. When administrators require iReady sessions, Khan loses time for actual science lessons.

Monitoring companies say schools can lock down devices through proper settings, yet the workarounds keep multiplying. Even young students now understand how to escape digital surveillance by using proxies or alternate accounts.

Parents Share Stories That Are Hard to Ignore

The stories coming from homes across LA are not abstract. They are visceral. One North Hollywood parent said her first grade son wet himself multiple times during mandatory iReady sessions because he was so overstimulated by headphones, animations, and game based quizzes that he ignored normal bodily signals. The incidents stopped only after the teacher reduced his device time.

Another parent described her teen becoming drawn into conversations with strangers online through the school’s tablet, eventually running away with the device.

Families who thought screen time would be manageable now find themselves wrestling with consequences they never anticipated.

A District Searching for Balance

District leaders continue to defend the device programs, saying they are essential to digital literacy. They argue that screens support learning when used responsibly and that schools should not restrict a tool that could help bridge socioeconomic divides.

Parents counter that the issue is not the technology itself but the scale and speed of its deployment. Many feel the classroom has tilted too far toward software driven instruction, with children plugged into devices for long stretches without human interaction.

Schools Beyond Screens now has chapters at roughly twenty schools. Their presence is growing. Their questions are straightforward. If children are learning less, behaving worse, and losing confidence, why are the screens still the center of instruction?

These are questions that do not demand ideology. They simply ask people to look at what is happening right in front of them.

What Happens Next?

The district plans to form a committee to review parent feedback next year. Some school board members are considering proposals that would forbid device use before second grade. Many parents are pushing for clearer oversight of software requirements and more transparency around the algorithms shaping their children’s education.

For now, Byock has opted her son out of iReady. His English teacher agreed to work with him one on one while the rest of the class uses the software. The improvement was immediate. The experience has left her with a belief that resonates with many parents across Los Angeles. Children learn best when educators see them, not their screens.

And families are beginning to ask why that has become such a radical idea.

Why Meta Shifted the Timeline

The delay followed a series of internal reviews. Business Insider reports that CEO Mark Zuckerberg told teams to take more time to strengthen the business case and ensure the product delivers a meaningfully improved experience. Updated memos from XR executives Gabriel Aul and Ryan Cairns said the additional months will give teams more room to refine the details and make sure the launch meets expectations.

This type of timeline reset is not new for Meta’s hardware efforts. The company has been pushing aggressively into AR and MR for several years, yet the market still lacks clear mainstream demand. When a device as important as Phoenix gets pushed back, it is often a signal that teams are adapting to realities that are not always visible from the outside.

Budget Shifts Add More Context

The delay also arrives just days after Bloomberg reported that Meta plans to reduce its metaverse budget by as much as 30 percent for 2026. A budget reduction of that scale does not necessarily indicate retreat. It does however speak to prioritization. XR hardware requires stable supply chains, careful pacing, and a clear narrative that can be communicated to users and investors. Meta has struggled with all three at different moments in recent years.

When a company adjusts both its timeline and budget in the same window, it usually means the broader strategy is evolving. For readers following the XR sector, these details are important. They help piece together a clearer picture of how Meta is navigating a market where innovation moves quickly but adoption moves slowly.

A Pattern Worth Paying Attention To

Mixed reality continues to hover between potential and practicality. Companies see the long term opportunity, but the path to a viable mass market device remains uneven. Meta’s delay is one more sign of this tension. Hardware can only advance as fast as the technology, economics, and user habits allow, and those elements do not always align.

This moment is a reminder to step back and observe the larger pattern. Product delays are rarely only about schedules. They often point to shifts in ambition, risk tolerance, or internal confidence. When patterns repeat, they tell a story long before the companies involved choose to spell it out.

Meta is still investing heavily in the future of AR and MR. The timeline adjustment for Project Phoenix simply positions the company to choose a path that is more measured and potentially more sustainable in a market that is still writing its own rules.

A Test Case for AI in Immigration Adjudications

The use of AI comes months after Acting EOIR Director Sirce E. Owen circulated Policy Memorandum 25‑40, which acknowledged that immigration courts have “neither a blanket prohibition on the use of generative AI in its proceedings nor a mandatory disclosure requirement regarding its use.”

The memo allows individual courts to adopt local rules on AI use but does not require judges to disclose its use. Burns appears to be among the first to adopt AI-assisted decision delivery, with courtroom assistants noting the technology has been in use since earlier this year.

Burns’ Record Raises Additional Concerns

Judge Burns is known as one of the most restrictive immigration judges in the U.S. Data compiled by Austin Kocher shows he approved just 2 percent of asylum claims between fiscal years 2019 and 2025, compared with a national average of 57.7 percent. Immigration advocates argue that pairing extreme denial rates with AI-assisted rulings may undermine defendants’ trust in the judicial process.

“When a person’s freedom depends on a judge’s voice, there needs to be certainty that it’s the judge’s own reasoning being rendered – not a synthesized layer of technology,” said an immigration lawyer who practices before the Broadway court.

Controversial Appointment and Political Pattern

Internal EOIR emails reveal that Burns’ appointment was politically influenced. Initially rated “Not Recommend” despite strong litigation experience, senior EOIR leadership overrode the evaluation and reclassified him as “Highly Recommended” in September 2020. Burns previously served as Assistant Chief Counsel for ICE in New York, representing the government in removal proceedings and appeals.

The pattern is part of a broader DOJ effort that has replaced more than 125 judges with politically aligned appointees and relaxed hiring standards, allowing the Attorney General to appoint “any licensed attorney” as a temporary immigration judge.

Broader Implications for Due Process

EOIR declined to comment specifically on Burns’ AI use. Experts warn that AI applications in courtrooms, when left undisclosed, accelerate the automation of adjudication in a system that already struggles with fairness and accountability.

“When the human element fades, so does accountability,” said a former EOIR official.

The incident underscores the need for clear oversight and transparency standards as generative AI enters sensitive legal proceedings.

Laterstack related stories here:

780,000 Windows users download a Linux distro in one month

AI data centers drive electricity prices higher and create political pressure across key states

Major Gainsight Supply Chain Breach Hits 200+ Companies

SAN FRANCISCO – Google has confirmed that hackers stole Salesforce-stored data from more than 200 companies in a large-scale supply chain hack via apps published by Gainsight, a customer support platform.

Austin Larsen, principal threat analyst at Google Threat Intelligence Group, stated that the company is “aware of more than 200 potentially affected Salesforce instances.”

The breach was claimed by the hacking collective Scattered Lapsus$ Hunters, which includes members of the ShinyHunters gang. The hackers stated they accessed data from major companies, including Atlassian, CrowdStrike, Docusign, F5, GitLab, LinkedIn, Malwarebytes, SonicWall, Thomson Reuters, and Verizon.

How the Hackers Gained Access

The hackers reportedly leveraged a prior campaign against Salesloft, which provides AI-driven marketing tools like Drift. They stole authentication tokens from Salesloft customers, allowing them to break into linked Salesforce instances through Gainsight.

“Gainsight was a customer of Salesloft Drift, they were affected and therefore compromised entirely by us,” said a spokesperson for ShinyHunters.

Salesforce confirmed the breach did not stem from vulnerabilities in its platform and has temporarily revoked active access tokens for Gainsight-connected apps as a precaution. Gainsight is collaborating with Google’s incident response unit, Mandiant, to investigate the incident.

Company Responses and Mitigation Efforts

Several affected companies provided statements on their current security posture:

CrowdStrike: Not affected; terminated a suspicious insider.

Docusign: No evidence of compromise; terminated Gainsight integrations and contained data flows.

Verizon & Thomson Reuters: Actively investigating.

Malwarebytes: Aware and investigating Gainsight/Salesforce issues.

Salesforce has stated it is following its policy of not commenting on specific customer cases.

Hacker Extortion Plans

The Scattered Lapsus$ Hunters collective has a history of data theft and extortion campaigns. Following the breach, they announced plans to launch a dedicated website to extort victims, mirroring tactics used in the October Salesloft incident. The group has previously targeted high-profile companies including MGM Resorts, Coinbase, and DoorDash.

These groups rely on social engineering and other techniques to compromise corporate systems, emphasizing the growing threat of sophisticated supply chain attacks in the cybersecurity landscape.

Looking Ahead

This incident highlights the risks of third-party software in enterprise environments, especially platforms connected to cloud services like Salesforce. Companies are advised to monitor unusual activity, enforce strict access controls, and coordinate with incident response teams to mitigate further exposure.