The dawn of the e-reader was a glorious moment for me after years of lugging around dozens of pounds of books to sate my bibliophile needs. Amazon‘s Kindle line stood out from its early, basic text form onward.
Now, Amazon has augmented its new Kindle Scribe with AI through a couple of very useful and surprisingly intuitive new features. Other companies making e-readers should take note, and conveniently, that’s exactly what the Kindle Scribe and its AI tools are built for.
In particular, while Amazon has marketed the Kindle Scribe as an E Ink notetaking device, the new Active Canvas facet of the e-reader lets you write notes on top of printed text, automatically gliding around and ensuring a sticky placement.
Chickenscratch refined
I’ll be the first to admit that my handwriting has never been the neatest. I’ve been told it’s perfect, but only for ransom notes and as a warning to children reluctant to practice their penmanship. It only worsens when I take notes quickly during a speech or interview. Trying to decipher it afterward is an art as much as a science, but the Kindle Scribe seems to have no trouble transforming handwritten notes, even messy ones, into legible text that’s much easier to read.
Advertisement
As someone who has always preferred jotting down notes by hand over typing or transcribing audio, that’s a huge deal. The AI keeps the charm of handwriting while keeping it useful. It’s a quiet deployment of AI, but a sign Amazon knows what Kindle Scribe users actually desire from AI tools.
From scattered to summarized
If you take a lot of notes, even having them be readable doesn’t mean you have them organized. That’s why the AI summarization feature for the new Kindle Scribe is so enticing. As a reporter, I might read and take notes on a PDF announcement for a new product, then go and take notes on the speech given by a company’s CEO when it is unveiled, and further write my comments on what I think about testing the product. The Kindle Scribe can distill those scattered notes written over many hours or days into a neat paragraph or two.
Indeed, the AI may not always extract the most relevant points from the notes. There might be extraneous bits left in or valuable data left out, but at least from what I’ve seen, that’s not a major issue with the Kindle Scribe’s AI. I would have cheerfully paid through the nose for such a feature when I was a student.
Sign up for breaking news, reviews, opinion, top tech deals, and more.
Advertisement
Again, Amazon is using AI in the Kindle Scribe to retain the appeal of digital note-taking while keeping things simple and streamlined. You don’t need an avalanche of options and a plethora of possibilities with AI for a digital reader and notebook. Enhancing the core writing and reading experiences with AI is no gimmick.
If the AI wearables struggling for sales this year had such obvious utility, they might not be struggling in the market. You might not think you need handwriting refinement and note summarization, but it’s hard to imagine giving them up once you start using them. Amazon’s AI may not be smarter than its rivals, but it certainly is employing it more intelligently in this case.
Connections is the latest puzzle game from the New York Times. The game tasks you with categorizing a pool of 16 words into four secret (for now) groups by figuring out how the words relate to each other. The puzzle resets every night at midnight and each new puzzle has a varying degree of difficulty. Just like Wordle, you can keep track of your winning streak and compare your scores with friends.
Some days are trickier than others. If you’re having a little trouble solving today’s Connections puzzle, check out our tips and hints below. And if you still can’t get it, we’ll tell you today’s answers at the very end.
In Connections, you’ll be shown a grid containing 16 words — your objective is to organize these words into four sets of four by identifying the connections that link them. These sets could encompass concepts like titles of video game franchises, book series sequels, shades of red, names of chain restaurants, etc.
There are generally words that seem like they could fit multiple themes, but there’s only one 100% correct answer. You’re able to shuffle the grid of words and rearrange them to help better see the potential connections.
Each group is color-coded. The yellow group is the easiest to figure out, followed by the green, blue, and purple groups.
Advertisement
Pick four words and hit Submit. If you’re correct, the four words will be removed from the grid and the theme connecting them will be revealed. Guess incorrectly and it’ll count as a mistake. You only have four mistakes available until the game ends.
We can help you solve today’s Connection by telling you the four themes. If you need more assistance, we’ll also give you one word from each group below.
Last week, the California Coastal Commission rejected a plan for SpaceX to launch up to 50 rockets this year at the Vandenberg Space Force Base in Santa Barbara County. The company responded yesterday with a lawsuit, alleging that the state agency’s denial was overreaching its authority and discriminating against its CEO.
The Commission’s goal is to protect California’s coasts and beaches, as well as the animals living in them. The agency has control over private companies’ requests to use the state coastline, but it can’t deny activities by federal departments. The denied launch request was actually made by the US Space Force on behalf of SpaceX, asking that the company be allowed to launch 50 of its Falcon 9 rockets, up from 36.
While the commissioners did raise about SpaceX CEO Elon Musk’s political screed and the spotty safety records at his companies during their review of the launch request, the assessment focused on the relationship between SpaceX and Space Force. The Space Force case is that “because it is a customer of — and reliant on — SpaceX’s launches and satellite network, SpaceX launches are a federal agency activity,” the Commission stated. “However, this does not align with how federal agency activities are defined in the Coastal Zone Management Act’s regulations or the manner in the Commission has historically implemented those regulations.” The California Coastal Commission claimed that at least 80 percent of the SpaceX rockets contain payloads for Musk’s Starlink company rather than payloads for government clients.
The SpaceX suit filed with the Central District of California court is seeking an order to designate the launches as federal activity, which would cut the Commission’s oversight out of its future launch plans.
Join our daily and weekly newsletters for the latest updates and exclusive content on industry-leading AI coverage. Learn More
Improving the capabilities of large language models (LLMs) in retrieving in-prompt information remains an area of active research that can impact important applications such as retrieval-augmented generation (RAG) and in-context learning (ICL).
Microsoft Research and Tsinghua University researchers have introduced Differential Transformer (Diff Transformer), a new LLM architecture that improves performance by amplifying attention to relevant context while filtering out noise. Their findings, published in a research paper, show that Diff Transformer outperforms the classic Transformer architecture in various settings.
Transformers and the “lost-in-the-middle” phenomenon
The Transformer architecture is the foundation of most modern LLMs. It uses an attention mechanism to weigh the importance of different parts of the input sequence when generating output. The attention mechanism employs the softmax function, which normalizes a vector of values into a probability distribution. In Transformers, the softmax function assigns attention scores to different tokens in the input sequence.
Advertisement
However, studies have shown that Transformers struggle to retrieve key information from long contexts.
“We began by investigating the so-called ‘lost-in-the-middle’ phenomenon,” Furu Wei, Partner Research Manager at Microsoft Research, told VentureBeat, referring to previous research findings that showed that LLMs “do not robustly make use of information in long input contexts” and that “performance significantly degrades when models must access relevant information in the middle of long contexts.”
Wei and his colleagues also observed that some LLM hallucinations, where the model produces incorrect outputs despite having relevant context information, correlate with spurious attention patterns.
“For example, large language models are easily distracted by context,” Wei said. “We analyzed the attention patterns and found that the Transformer attention tends to over-attend irrelevant context because of the softmax bottleneck.”
Advertisement
The softmax function used in Transformer’s attention mechanism tends to distribute attention scores across all tokens, even those that are not relevant to the task. This can cause the model to lose focus on the most important parts of the input, especially in long contexts.
“Previous studies indicate that the softmax attention has a bias to learn low-frequency signals because the softmax attention scores are restricted to positive values and have to be summed to 1,” Wei said. “The theoretical bottleneck renders [it] such that the classic Transformer cannot learn sparse attention distributions. In other words, the attention scores tend to flatten rather than focusing on relevant context.”
Differential Transformer
To address this limitation, the researchers developed the Diff Transformer, a new foundation architecture for LLMs. The core idea is to use a “differential attention” mechanism that cancels out noise and amplifies the attention given to the most relevant parts of the input.
The Transformer uses three vectors to compute attention: query, key, and value. The classic attention mechanism performs the softmax function on the entire query and key vectors.
The proposed differential attention works by partitioning the query and key vectors into two groups and computing two separate softmax attention maps. The difference between these two maps is then used as the attention score. This process eliminates common noise, encouraging the model to focus on information that is pertinent to the input.
Advertisement
The researchers compare their approach to noise-canceling headphones or differential amplifiers in electrical engineering, where the difference between two signals cancels out common-mode noise.
While Diff Transformer involves an additional subtraction operation compared to the classic Transformer, it maintains efficiency thanks to parallelization and optimization techniques.
“In the experimental setup, we matched the number of parameters and FLOPs with Transformers,” Wei said. “Because the basic operator is still softmax, it can also benefit from the widely used FlashAttention cuda kernels for acceleration.”
In retrospect, the method used in Diff Transformer seems like a simple and intuitive solution. Wei compares it to ResNet, a popular deep learning architecture that introduced “residual connections” to improve the training of very deep neural networks. Residual connections made a very simple change to the traditional architecture yet had a profound impact.
Advertisement
“In research, the key is to figure out ‘what is the right problem?’” Wei said. “Once we can ask the right question, the solution is often intuitive. Similar to ResNet, the residual connection is an addition, compared with the subtraction in Diff Transformer, so it wasn’t immediately apparent for researchers to propose the idea.”
Diff Transformer in action
The researchers evaluated Diff Transformer on various language modeling tasks, scaling it up in terms of model size (from 3 billion to 13 billion parameters), training tokens, and context length (up to 64,000 tokens).
Their experiments showed that Diff Transformer consistently outperforms the classic Transformer architecture across different benchmarks. A 3-billion-parameter Diff Transformer trained on 1 trillion tokens showed consistent improvements of several percentage points compared to similarly sized Transformer models.
Further experiments with different model sizes and training dataset sizes confirmed the scalability of Diff Transformer. Their findings suggest that in general, Diff Transformer requires only around 65% of the model size or training tokens needed by a classic Transformer to achieve comparable performance.
Advertisement
The researchers also found that Diff Transformer is particularly effective in using increasing context lengths. It showed significant improvements in key information retrieval, hallucination mitigation, and in-context learning.
While the initial results are promising, there’s still room for improvement. The research team is working on scaling Diff Transformer to larger model sizes and training datasets. They also plan to extend it to other modalities, including image, audio, video, and multimodal data.
The researchers have released the code for Diff Transformer, implemented with different attention and optimization mechanisms. They believe the architecture can help improve performance across various LLM applications.
“As the model can attend to relevant context more accurately, it is expected that these language models can better understand the context information with less in-context hallucinations,” Wei said. “For example, for the retrieval-augmented generation settings (such as Bing Chat, Perplexity, and customized models for specific domains or industries), the models can generate more accurate responses by conditioning on the retrieved documents.”
VB Daily
Advertisement
Stay in the know! Get the latest news in your inbox daily
Zepto is in advanced stages of talks to raise $100 million in new investment, its third in the last six months, as the leading Indian quick commerce startup looks to rope in more domestic investors, sources familiar with the talks told TechCrunch.
The Mumbai-headquartered startup, which delivers grocery items and office stationery to customers’ doorsteps in 10 minutes in multiple Indian cities, is raising the new investment from Indian family offices and high net worth individuals.
Motilal Oswal, the asset management giant that earlier invested $40 million in Zepto, is running the mandate for the new funding deliberation, the sources said, requesting anonymity as the matter is private. The financial services firm has already received commitments for more than half of the allocation, according to another source familiar with the situation.
The new investment values Zepto at a $5 billion post-money valuation, the same value at which it recently closed a $340 million financing round in August. Zepto has raised more than $1 billion in the last six months and all of it remains in its bank.
Advertisement
Zepto is planning to go public next year and the new fundraise is aimed at expanding the base of domestic investors on its cap table. Zepto counts Avra, Lightspeed, Nexus, StepStone Group, YC Continuity, Glade Brook and Contrary among its backers.
Even as quick commerce startups are retreating, consolidating or shutting down in many parts of the world, the model is showing encouraging signs in India. Quick commerce startups are on track to do a sale of more than $6 billion this year, according to TechCrunch’s analysis.
In response to the fast rise of quick commerce, which is increasingly shaping the consumer behavior in India, many e-commerce incumbents — including Flipkart, Myntra and Nykaa have been forced to scramble ways to lower the time they take to deliver items to their customers.
Shares of Dmart, which runs one of the largest brick-and-mortar retail chains in India, fell this week after the firm confirmed that it was losing some business to quick commerce startups.
Advertisement
“We believe Quick Commerce players are expanding cities, categories, SKUs, AOVs and discounts, and creating parallel commerce for convenience-seeking customers,” analysts at Morgan Stanley wrote in a note this week.
Zepto – which competes with Zomato-owned BlinkIt, Prosus-backed Swiggy’s Instamart, and Tata’s BigBasket – has grown its annualized net runrate considerably in recent months, according to sources and an internal document reviewed by TechCrunch.
DJI tells The Verge that it currently cannot freely import all of its drones into the United States — and that its latest consumer drone, the Air 3S, won’t currently be sold at retail as a result.
“A customs-related issue is hindering DJI’s ability to import select drones into the United States.”
That’s not because the United States has suddenly banned DJI drones — rather, DJI believes the import restrictions are “part of a broader initiative by the Department of Homeland Security to scrutinize the origins of products, particularly in the case of Chinese-made drones,” according to DJI.
DJI recently sent a letter to distributors with one possible reason why DHS is stopping some of its drones: the company says US Customs and Border Protection is citing the Uyghur Forced Labor Prevention Act (UFLPA) as justification for blocking the imports. In the letter, which has been floating around drone sites and Reddit for several days, DJI claims it doesn’t use any forced labor to manufacture drones.
Advertisement
Reuters reported on the letter earlier today; DJI spokesperson Daisy Kong confirmed the letter’s legitimacy to The Verge as well.
In a just-published official blog post, DJI is calling this all a “misunderstanding,” and writes that it’s currently sending documentation to US Customs to prove that it doesn’t manufacture anything in the Xinjiang region of China where Uyghurs have been forcibly detained, that it complies with US law and international standards, and that US retailers have audited its supply chain. DJI claims it manufacturers all its products in Shenzhen or Malaysia.
US Customs and Border Protection didn’t reply to a request for comment.
While the US House of Representatives did pass a bill that would effectively ban DJI drones from being imported into the US, that ban would also need to pass the Senate. Last we checked, the Senate had removed the DJI ban from its version of the must-pass 2025 National Defense Authorization Act (though it did get reintroduced as an amendment and could potentially still make it into the final bill).
Advertisement
DJI says the “customs-related issue” has “primarily impacted” the company’s enterprise and agricultural drones, but has also now “limited us from offering the Air 3S to US customers beyond DJI.com.”
“We are actively working with U.S. Customs and Border Protection to resolve this issue and remain hopeful for a swift resolution,” writes DJI.
The US government has cracked down on DJI drones before, but not in a way that would keep stores from buying them, consumers from purchasing them, or individual pilots from flying them in the United States. Primarily, the US Department of Commerce’s “entity list” keeps US companies from exporting their technology to the Chinese company, and the US has sometimes restricted certain government entities from purchasing new DJI drones.
Even if DJI imports do get banned by Congress, the proposed law suggests existing owners could still use their drones — but the FCC could no longer authorize DJI gadgets with radios for use in the United States, which would effectively block all imports.
A whopping 69% of organizations have reported paying ransoms this year, according to research by Cohesity, with 46% handing over a quarter of a million dollars or more to cybercriminals. It is hardly the picture of resiliency that is often painted by industry. Clearly, there is a disconnect between cyber resiliency policy and operational capability that urgently needs addressing.
With the advent of Ransomware-as-a-Service platforms and the current global geopolitical situation, organizations face a huge existential threat through destructive cyber attacks that could put them out of business. This gap between confidence and capability needs to be addressed, but in order to do so, those organizations need to recognize there is a problem in the first place.
According to the Global cyber resilience report 2024, which surveyed 3,139 IT and Security Operations (SecOps) decision-makers, despite 77% of companies having a ‘do not pay’ policy, many have found themselves unable to respond and recover from attacks without caving in to ransom demands. In addition, only 2% of organizations can recover their data and restore business operations within 24 hours of a cyberattack – despite 98% of organizations claiming their recovery target was one day.
This clearly indicates that current cyber resilience strategies are failing to deliver when it matters most. Companies have set ambitious recovery time objectives (RTOs), but are nowhere close to building the appropriate effective and efficient investigation and threat mitigation capability needed to rebuild and recover securely. Most organizations treat a destructive cyber attack like a traditional business continuity incident like a flood, fire or electricity loss – recovering from the last backup and bringing back in all the vulnerabilities, gaps in prevention and detection, as well as persistence mechanisms that caused the incident in the first place. The gap between these goals and actual capabilities is a ticking time bomb, leaving businesses vulnerable to prolonged downtime and severe financial losses.
Equally alarming is the widespread neglect of Zero-Trust Security principles. While many companies tout their commitment to securing sensitive data, less than half have implemented multi-factor authentication (MFA) or role-based access controls (RBAC). These are not just best practices; they are essential safeguards in today’s threat landscape. Without them, organizations are leaving the door wide open to both external and internal threats.
Advertisement
As cyber threats continue to evolve, with 80% of companies now facing the threat of AI-enabled attacks, the need for a robust, modern approach to data resiliency is more urgent than ever. Yet, the continued reliance on outdated strategies and the failure to adapt to new threats sets the stage for even greater risks. It’s not even a question of complacency.
James Blake
Global Head of Cyber Resiliency Strategy at Cohesity.
Building confidence or creating false hope?
With 78% of organizations claiming that they are confident in their cyber resilience capability, this infers that a lot of work has already been done in creating the process and technology to not just isolate attacks but also have the ability to recover a trusted response capability to investigate, mitigate threats and recover. This would be great if true, but we are seeing a real disconnect between perception and reality when it comes to cyber resilience.
Advertisement
Sign up to the TechRadar Pro newsletter to get all the top news, opinion, features and guidance your business needs to succeed!
That’s a big concern. The financial impact of these failures is not limited to ransom payments alone. The true cost of inadequate cyber resilience extends far beyond the immediate outlay. Prolonged downtime, loss of customer trust, criminal prosecutions for false attestations around the quality of security controls or paying ransoms to sanctioned entities, brand damage, and skyrocketing cyber insurance premiums are just a few consequences that can damage an organization. It’s a sobering reminder that investing in and testing robust cyber resiliency measures upfront is far more cost-effective than dealing with the fallout of a successful attack.
Moreover, the report reveals that only 42% of organizations have the IT and Security capabilities to identify sensitive data and comply with their regulatory requirements. This deficiency exposes companies to significant fines and undermines their ability to prioritize protecting the very data that is the lifeblood of their organization and is subject to regulatory obligations.
With the expected rise of AI-enhanced cyberattacks adding another layer of capability to cyber adversaries, organizations with traditional defenses will have their work cut out. They are no match for these effective and high-efficient threats, which can adapt and evolve faster than most organizations can respond. Organizations need AI-tools to counter these emerging AI-driven threats.
Advertisement
Identify a problem to fix a problem
The report ultimately reveals opportunities for improvement. People, processes, and tools do exist to reverse these trends and close gaps to shore up cyber resilience. Still, organizations need to understand where they currently sit regarding resiliency and be honest with themselves.
The right workflow collaboration and platform integration between IT and Security needs to be developed before an incident. Organizations must engage in more realistic and rigorous threat modelling, attack simulations, drills and tests to understand their strengths and weaknesses. This can ensure that the response and recovery process is effective and that all stakeholders are familiar with their roles during an incident or can identify shortcomings and areas for improvement.
In addition, automated testing of backup data can verify the integrity and recoverability of backups without manual intervention. This automation helps ensure that backups are reliable and can be restored quickly when needed.
Finally, maintaining detailed documentation and recovery playbooks helps ensure everyone knows their responsibilities and what steps to take during an incident. These playbooks should be regularly updated based on changes in adversary behavior and the results of testing and drills.
Advertisement
And this is just a start. To fully reduce operational risk, a transition to modern data security and management processes, tools, and practices is required. Perhaps then, we will see a reduction in ransom payments and a cyber resilience confidence built on reality.
This article was produced as part of TechRadarPro’s Expert Insights channel where we feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/news/submit-your-story-to-techradar-pro
You must be logged in to post a comment Login