gHacks Technology News

Prompted Playlist Lets Users Describe The Music They Want

Spotify has unveiled a new AI-powered feature called Prompted Playlist that generates playlists based on written descriptions. Instead of selecting genres or artists, users can describe a vibe, scenario, or feeling, and Spotify builds a playlist around that prompt.

The feature also factors in the user’s listening history and current music trends, combining personalization with real-time data from the platform.

How Prompted Playlist Works Playlists Generated From Written Prompts

Prompted Playlist allows users to type a short description such as a mood, moment, or cultural reference. Spotify then creates a playlist designed to match that idea. The system can also avoid recently overplayed tracks and incorporate artists tied to current pop culture trends.

Spotify says early testing showed users creating playlists for specific moments, such as late-night listening or seasonal moods, while filtering out songs they had heard too often.

Editable And Refreshable Playlists

Once generated, playlists can be adjusted by editing the original prompt. Users can also set playlists to refresh automatically on a daily or weekly basis, keeping selections current without needing to start over.

If a Prompted Playlist is shared, other listeners receive their own personalized version rather than a static copy.

Availability And Requirements

Prompted Playlist is currently rolling out in beta to Spotify Premium subscribers in the United States and Canada. The feature was previously tested in New Zealand before expanding to additional markets.

The tool is accessed by selecting Create in the Spotify app and choosing Prompted Playlist. From there, users enter a prompt and let Spotify generate the playlist automatically.

Part Of A Broader AI Push At Spotify

Prompted Playlist follows earlier experiments with AI-generated playlists and other personalization features. Spotify says the goal is to shift users from passively receiving recommendations to actively guiding the algorithm using their own words.

The rollout comes shortly after Spotify announced subscription price increases in several regions, including a $1 increase for Premium plans in the United States beginning in February.

Spotify has not said when or if Prompted Playlist will expand beyond beta or become available to free users, but the company continues to position AI-driven personalization as a core part of its platform strategy.

Thank you for being a Ghacks reader. The post Spotify Introduces Prompted Playlist AI Feature appeared first on gHacks Technology News.

BBC Shifts YouTube From Promotion To Primary Distribution

BBC has announced a new partnership with YouTube that will see the broadcaster create original programmes designed specifically for YouTube. The move marks a change in strategy, as the BBC has historically used YouTube mainly to host clips and trailers that point viewers back to its own services.

Under the new agreement, YouTube will act as the primary destination for selected programmes. Some content may also be made available on BBC-owned platforms such as BBC iPlayer and BBC Sounds, but the projects are being commissioned with YouTube audiences in mind from the outset.

Advertising And Revenue Outside The UK

One practical outcome of the partnership is monetisation. While BBC content on YouTube will remain ad-free for viewers in the UK, advertising will be enabled for international audiences. This creates an additional revenue stream at a time when the BBC’s funding model and licence fee arrangements continue to face scrutiny.

The agreement covers both UK public service output and international content produced by BBC Studios.

What Kind Of Content To Expect

Programming developed under the partnership is expected to span multiple genres, including entertainment, documentaries, children’s content, news, and sport. Coverage tied to the Winter Olympics is among the first projects currently planned.

The BBC says the goal is to reach younger, digital-first audiences who increasingly consume video through social and creator-driven platforms rather than traditional broadcast channels.

BBC And YouTube On The Partnership

In a joint statement, the BBC and YouTube said the deal would highlight “the best homegrown storytelling and news from across the BBC.” BBC Director General Tim Davie described the move as an expansion of the broadcaster’s digital ambitions, noting that the partnership also creates new pathways for audiences to discover BBC services beyond YouTube.

YouTube’s EMEA leadership framed the deal as a way to adapt the BBC’s content to platform-native formats while also investing in creator training within the UK.

A Larger Shift In Viewing Habits

The BBC already operates multiple high-profile YouTube channels, including a main account focused on clips and trailers and a dedicated BBC News channel, together reaching tens of millions of subscribers. The broadcaster is expected to expand its YouTube presence significantly as part of the new strategy.

The move comes as viewing data continues to show YouTube competing directly with traditional broadcasters for audience attention, particularly among younger viewers. The BBC’s decision to commission YouTube-first programmes reflects that shift, positioning the platform as more than a secondary outlet for broadcast content.

Thank you for being a Ghacks reader. The post BBC Partners With YouTube To Produce Original Platform-First Content appeared first on gHacks Technology News.

Siri May Shift Toward Conversational AI

Apple is reportedly preparing a major change to Siri that would make the assistant behave more like a modern AI chatbot. According to reporting from Mark Gurman, Apple is developing a new Siri interface designed around continuous conversations rather than single, one-off commands.

The change would represent a shift from Siri’s current interaction model, which remains largely command-based despite recent AI additions. Apple has already acknowledged delays in delivering deeper contextual features promised in earlier updates.

What A Chatbot-Style Siri Would Change A Persistent Conversation Interface

The new Siri interface is said to resemble dedicated chatbot apps such as ChatGPT or Google’s Gemini. Users would be able to type or speak to Siri and move seamlessly between both input methods, with conversation history preserved and scrollable.

While users can already type requests to Siri, those interactions do not currently function as an ongoing dialogue. Siri cannot reference earlier requests or maintain context over time, which puts it behind competing AI assistants.

Deeper Integration Across Apple Apps

Beyond interface changes, the chatbot-style Siri is expected to be integrated more broadly across Apple’s core apps. That includes Messages, Mail, Music, Photos, Podcasts, TV, and developer tools like Xcode. This would give Siri access to more on-device context than most third-party chatbots, though it also raises questions around scope and control.

The new interface would build on personalization and contextual features Apple has already announced, rather than replacing them.

Timeline And Platform Details Expected To Arrive With Future OS Releases

The chatbot-style Siri is reportedly planned for future versions of iOS and macOS, internally referred to as part of a project named “Campos.” Gurman suggests that Apple may share more details at WWDC, which typically takes place in June.

The contextual Siri upgrades previously announced are expected to arrive earlier, with the chatbot interface following later as a separate phase.

How This Fits Into Apple’s AI Strategy

Apple has previously downplayed the idea of turning Siri into a standalone chatbot, positioning its AI efforts around tightly integrated system features rather than chat-first experiences. A conversational Siri interface suggests that stance may be evolving, even if Apple continues to frame the change as part of Apple Intelligence rather than a direct chatbot replacement.

The assistant is still expected to rely on external AI models, including Google’s Gemini, rather than Apple’s own large language models.

If implemented as described, the new Siri would not compete on openness or extensibility but could become the default AI assistant for iPhone and Mac users due to its deep system access. Whether Apple delivers the feature as outlined remains unconfirmed, and the company has not publicly acknowledged the reported plans.

Thank you for being a Ghacks reader. The post Apple Plans A Chatbot-Style Siri Interface appeared first on gHacks Technology News.

Apple Turns To Google Gemini For Apple Intelligence

Apple has confirmed that upcoming Apple Intelligence features on iPhone will be powered in part by Google’s Gemini AI models. The change follows delays to Apple’s own AI roadmap and marks a significant shift in how Siri and system-level intelligence will evolve over the next year.

While Apple has not published a full feature list, multiple reports outline what Gemini-backed capabilities are expected to arrive across upcoming iOS releases.

Siri Features Expected To Use Gemini Models More Conversational And Context-Aware Responses

A revamped version of Siri is expected to deliver more natural, chatbot-style responses. Instead of short, transactional replies, Siri will be able to provide longer conversational answers similar to ChatGPT and Gemini. This includes improved handling of ambiguous questions and follow-up prompts when requests are unclear.

Expanded Task Handling

Gemini-powered Siri is expected to perform more complex actions, including:

• Creating documents directly in apps like Apple Notes
• Booking travel and handling multi-step requests
• Taking action inside apps rather than redirecting users

These changes build on Apple’s previously announced goal of making Siri more useful across the system rather than acting as a voice-only shortcut tool.

Emotional And Storytelling Responses

Reports also suggest Siri will be capable of generating stories and offering supportive responses when users express emotional states. Apple has not clarified how these features will be presented or controlled, but they are part of the broader conversational upgrade.

Apple Intelligence Features Revived With Gemini Support Personal Context And On-Screen Awareness

Several Apple Intelligence features first shown at WWDC 2023 were delayed and are now expected to ship using Gemini models. These include:

• Personal context awareness to locate messages, emails, or files stored on the device
• On-screen awareness that allows Siri to interact with visible content
• In-app actions that let users issue commands specific to active apps

These capabilities rely on deeper system access than third-party AI apps typically receive.

Features Planned For Future iOS Releases Memory And Proactive Suggestions

Looking further ahead, Siri features expected to debut with future iOS versions include the ability to remember past conversations and provide proactive suggestions. Examples include recommending when to leave for an appointment based on traffic or reminding users about upcoming plans without being prompted.

Apple is expected to share more details at WWDC, traditionally held in June.

When These Features Are Expected To Arrive

The first Gemini-powered features are reportedly on track to arrive in an upcoming iOS point release, with broader Siri upgrades following in later versions. Apple has not confirmed exact timelines, but beta testing is expected to begin soon.

If delivered as outlined, Gemini integration would represent the largest functional upgrade to Siri in years, shifting it closer to a full conversational assistant while remaining deeply embedded within Apple’s ecosystem.

Thank you for being a Ghacks reader. The post New iPhone Features Powered By Google Gemini Are Coming Soon appeared first on gHacks Technology News.

Notepad Gains Expanded Markdown And Faster Text Generation

Microsoft has revealed a set of upcoming updates for Windows 11 built-in apps, with notable improvements coming to Notepad.

Notepad is receiving expanded Markdown support. New syntax options include strikethrough formatting and nested lists, building on earlier Markdown features. Users can continue to rely on keyboard shortcuts, but Microsoft has also added toolbar controls for applying formatting without typing Markdown manually.

Another visible change is a new Welcome Experience. This popup appears when Notepad receives feature updates and highlights what has changed. With Notepad now updated more frequently than in the past, the screen is intended to help users keep track of new capabilities without scanning release notes.

Microsoft is also improving AI-powered text generation in Notepad. Generated text will now appear using a streaming effect, displaying responses as they are created instead of waiting for the full output to complete. The change is designed to make results feel faster, even when generation time stays the same.

Paint Adds AI Coloring Book Mode And Better Fill Control AI Coloring Book For Copilot+ PCs

Updates are also coming to Microsoft Paint. The most eye-catching addition is a new AI-powered Coloring Book mode. Available on Copilot+ PCs, the feature uses the device’s NPU to generate outline-style drawings based on text prompts.

Users can open Paint, select the Coloring Book option from the Copilot menu, and describe the scene they want. Paint then generates a set of black-and-white outlines designed to be colored in, turning simple prompts into printable or editable coloring pages.

Adjustable Fill Tolerance

Paint is also gaining a fill tolerance control. This lets users decide how aggressively the fill tool applies color within a selected area. Lower tolerance values prevent color from bleeding into nearby regions, making the tool more useful for detailed drawings and outlines.

Availability And Rollout

The new Notepad and Paint features are rolling out first to Windows Insiders in the Canary and Dev channels. Microsoft says they will reach all Windows 11 users in the coming weeks, with the Coloring Book mode limited to supported Copilot+ PCs.

Users interested in the Markdown changes or new Paint tools should check app updates regularly, as these improvements are being delivered through app updates rather than full Windows feature releases.

Thank you for being a Ghacks reader. The post Microsoft Previews New Windows 11 Features For Notepad And Paint appeared first on gHacks Technology News.

PlayStation Plus Lineup Shrinks Again Next Month

Sony will remove nine games from PlayStation Plus Extra and Premium tiers in February 2026. As with previous monthly rotations, the removals will happen on the same day new titles are added to the service.

The affected games will no longer be playable for subscribers who have not purchased them individually once they leave the catalog.

Games Leaving PlayStation Plus On February 17

The following titles are scheduled to leave PlayStation Plus Extra and Premium on February 17, 2026:

Full List Of Departing Games

• Carto
• Cult of the Lamb
• New World: Aeternum
• Rez Infinite
• SaGa Frontier Remastered
• Spirit of the North: Enhanced Edition
• Super Neptunia RPG
• The Ascent
• WWE 2K25

One notable case is New World: Aeternum, which has already been delisted ahead of its planned server shutdown in January 2027. Subscribers who did not download it earlier can no longer access it through PlayStation Plus.

Notable Removals And Time-Limited Access

Several of the departing games are considered highlights of the Extra and Premium libraries. Cult of the Lamb earned strong critical reception for its blend of roguelike combat and management elements, while Rez Infinite has been one of the more prominent VR-capable titles available through the service.

WWE 2K25 is also leaving after a relatively short stay. While the game remains playable offline, its online servers are scheduled to shut down in March 2027, which may factor into whether subscribers want to spend time with it before removal.

The Ascent offered local co-op gameplay, and its departure reduces the number of couch co-op options available through PlayStation Plus.

What Subscribers Should Do Now

Subscribers who want to finish or revisit any of these titles will need to do so before February 17. Once removed, access will be revoked unless the games are purchased separately from the PlayStation Store.

Sony is expected to announce the February 2026 additions to PlayStation Plus Extra and Premium a few days before the rotation, following its usual schedule.

Thank you for being a Ghacks reader. The post Nine Games Are Leaving PlayStation Plus Extra And Premium In February 2026 appeared first on gHacks Technology News.

OpenAI Expands Its Government-Focused Education Initiative

OpenAI has introduced Education for Countries, a new pillar under its broader OpenAI for Countries initiative. The program is designed to help governments modernize education systems and prepare future workforces by embedding AI tools, training, and research directly into national education infrastructure.

The initiative targets what OpenAI describes as a growing gap between what modern AI systems can do and how widely they are used in everyday learning and work. Education systems are positioned as a primary way to close that gap, particularly as AI-driven changes reshape job requirements.

What Education For Countries Includes 1. AI Tools Integrated Into Learning Systems

Participating countries gain access to AI tools such as ChatGPT Edu, newer GPT models, study mode, and collaborative features like canvas. These tools can be customized to reflect local curricula, language needs, and education priorities rather than being deployed as generic platforms.

2. Research On Learning Outcomes And Productivity

The program includes partnerships focused on large-scale education research. These studies examine how AI affects student learning, teacher workload, and institutional productivity. Results are intended to inform national policy decisions, workforce planning, and future AI development.

3. Training And Certifications For Educators And Students

OpenAI is also providing structured training through initiatives such as the OpenAI Academy and AI-focused certifications. These programs are designed to align practical AI skills with national workforce goals, targeting both educators and students.

4. A Network Of Government And Academic Partners

Education for Countries brings together governments, universities, and researchers into a shared network. Participants are expected to exchange deployment experiences, share best practices, and collaborate on responsible AI use in education.

Countries Participating In The First Cohort

The initial group includes Estonia, Greece, Italy (through the Conference of University Rectors), Jordan, Kazakhstan, Slovakia, Trinidad and Tobago, and the United Arab Emirates.

Estonia is highlighted as an early example. ChatGPT Edu has already been deployed nationwide across public universities and secondary schools, reaching more than 30,000 students, educators, and researchers in its first year. Long-term studies are also underway, including research partnerships measuring AI’s impact on learning outcomes across tens of thousands of students.

How AI Is Being Rolled Out In Schools

OpenAI describes a phased rollout model. Educators are typically equipped first, receiving tools and training before broader student access is introduced. In higher education, students already have direct access to ChatGPT Edu. In secondary education, student access begins with limited pilots developed in coordination with local authorities.

These pilots are paired with safeguards aimed at younger users, including age-appropriate model behavior and AI literacy resources developed with partners such as Common Sense Media.

Broader Context And Next Steps

Education for Countries builds on OpenAI’s existing education-focused efforts, including certification programs and university research partnerships. The company frames the initiative as part of its broader mission to ensure advanced AI systems expand opportunity rather than limit access.

OpenAI has stated that additional countries will join the program in future cohorts, with the next group expected to be announced later in 2026.

Thank you for being a Ghacks reader. The post OpenAI Expands AI Education Programs At A National Level appeared first on gHacks Technology News.

CursorWrap Adds Edge Wrapping For Mouse Navigation

Microsoft has released PowerToys version 0.97.0, introducing a new utility called CursorWrap. The feature is aimed primarily at users with multi-monitor setups, though it can also be useful on large single displays.

CursorWrap allows the mouse pointer to wrap around screen edges. Moving the cursor past the right edge of the right-most display causes it to reappear on the left edge of the left-most display. The same behavior applies to the top and bottom edges.

The utility includes several configuration options. Users can toggle CursorWrap on or off with a keyboard shortcut and disable wrapping while dragging items, preventing accidental repositioning during drag-and-drop actions. Recent improvements also limit wrapping to the outer edges of multi-monitor layouts, avoiding unexpected behavior between internal monitor boundaries.

PowerToys v0.97.0 Includes Broad Improvements Across Modules Command Palette Sees Major Expansion

PowerToys v0.97.0 delivers a large number of updates beyond CursorWrap, with Command Palette receiving the most attention. The release adds drag-and-drop support, new navigation controls, improved result ranking, theme customization, and a built-in Remote Desktop extension. Performance and accessibility have also been improved, and the Command Palette version has been bumped to 0.8.

New Features And Fixes Across Utilities

Other notable changes include a new screenshot-freezing mode in Crop & Lock, a command-line interface for FancyZones and File Locksmith, and better handling of mixed-DPI setups. Image Resizer, PowerRename, Quick Accent, and Find My Mouse all received fixes or small feature additions.

Microsoft also continued refactoring older modules such as Mouse Without Borders and added more telemetry hooks across several tools to better understand usage patterns.

Settings And Development Changes

The Settings interface now includes a new adaptive tray icon, improved accessibility headings, and simplified internal logic. On the development side, Microsoft updated build tooling, documentation, and testing pipelines, focusing on stability and long-term maintenance.

Availability

PowerToys v0.97.0 is available now and can be installed via the built-in updater or downloaded manually from Microsoft’s release channels. Users interested in CursorWrap or Command Palette enhancements should review the new settings after updating, as many features are configurable but not enabled by default.

Thank you for being a Ghacks reader. The post Microsoft Releases PowerToys v0.97.0 With New CursorWrap Utility appeared first on gHacks Technology News.

Spotify Is Experimenting With Page-Based Audiobook Syncing

Spotify appears to be working on a new audiobook feature designed to bridge physical books and audiobooks. Code discovered in recent versions of the app points to a system that can sync audiobook progress with printed books or e-books by scanning pages using a smartphone camera.

The feature, currently referred to as Page Match in internal code, was uncovered by a tipster and is not yet available to users. Spotify has not officially announced it, but the references suggest active development rather than a dormant experiment.

How Page Match Is Designed To Work Scan A Page To Sync Audiobook Progress

Page Match is built around user interaction with a physical book. When reading a printed copy, users would be able to scan a page using their phone. Spotify would then adjust the audiobook playback position to match that page, allowing listening to resume from the same point.

This approach removes the need to manually scrub through an audiobook to find the correct chapter or timestamp. It is designed for users who switch between reading and listening rather than sticking to one format.

Reverse Syncing From Audio To Print

The code also suggests Page Match can work in reverse. If a user has been listening to an audiobook, Spotify may indicate which page they are currently on, allowing them to continue reading from the correct spot in the physical or digital book.

This two-way syncing hints at deeper integration between audiobook metadata and text-based editions, assuming Spotify has access to accurate page mapping data.

Availability And Limitations

Spotify has not provided a release timeline for Page Match. The feature is still under development, and there is no confirmation that it will ship in its current form.

Based on how Spotify handles audiobooks today, Page Match is expected to be limited to Premium subscribers and only available in regions where Spotify already offers audiobooks. Markets without audiobook support are unlikely to receive the feature if it launches.

Spotify’s audiobook catalog currently operates under a time-based model, with Premium users receiving a monthly listening allowance before additional charges apply. Page Match would build on that system rather than replace it.

Recent Spotify Changes Provide Context

Spotify has been expanding beyond music over the past few years, investing heavily in podcasts and audiobooks. The company recently added social features such as Request to Jam in messages and increased U.S. subscription prices across several plans.

Page Match fits into that broader strategy by making audiobooks more flexible rather than simply increasing catalog size. Whether it becomes a widely used feature depends on accuracy, regional availability, and whether users are willing to scan pages as part of their listening workflow.

For now, Page Match remains an unannounced feature under development, with no confirmation on when or if it will be released publicly.

Are you a Spotify user? Or prefer Apple Music?

Thank you for being a Ghacks reader. The post Spotify Is Testing A Feature That Syncs Physical Books With Audiobooks appeared first on gHacks Technology News.

A set of recent changes to Wine could remove one of Linux’s longest-standing software roadblocks. A developer working under the name PhialsBasement has introduced patches that allow Adobe Creative Cloud to install on Linux, making it possible to run Adobe Photoshop through Wine.

The work targets failures that have historically stopped Adobe’s installer from running at all. Creative Cloud relies on Windows components that Wine does not fully implement, causing the setup process to abort early. The new patches focus on closing those gaps rather than modifying Adobe software directly.

According to the developer, two missing pieces have been key blockers. One is MSHTML, Microsoft’s legacy browser engine used by some Windows applications for embedded web interfaces. The other is MSXML, a set of XML services required by many script-driven installers and applications on Windows XP SP3 and later. Without these components behaving correctly, Creative Cloud cannot complete its setup routine.

The patches introduce support for these dependencies in Wine, allowing the Creative Cloud installer framework to run as expected. With the changes applied, both Photoshop 2021 and Photoshop 2025 were reported to install successfully and launch on Linux systems. Basic functionality works, though the setup is not described as flawless.

Other users testing the same code have confirmed that compatibility is improved, but limitations remain. Performance issues, missing features, and occasional glitches are still present, and the experience does not yet match running Photoshop on Windows or macOS. The changes focus on enabling installation and execution rather than polishing day-to-day usability.

The patches were initially submitted to Valve’s custom Wine tree, used for projects like Proton. The developer has since been asked to resubmit the work to Wine’s main repository on GitLab, where it would need to go through review before being accepted upstream. That process can take time, and acceptance is not guaranteed.

Wine already supports a large number of Windows applications, particularly games, but creative software has been a weaker area. Adobe applications are among the most frequently cited reasons professionals avoid Linux on the desktop. Improving Creative Cloud compatibility addresses a practical limitation rather than adding a niche feature.

Adobe Creative Cloud remains a subscription-based platform that manages applications, updates, and licensing. Photoshop continues to be treated as an industry standard for image editing, which makes its absence on Linux more noticeable than many other Windows-only tools.

At this stage, the work remains experimental and limited to custom Wine builds. Users interested in testing it should expect manual setup and incomplete behavior. Whether the patches are merged into upstream Wine will determine how accessible Photoshop on Linux becomes in the future.

Thank you for being a Ghacks reader. The post Wine Patches Enable Adobe Photoshop Installation On Linux appeared first on gHacks Technology News.

Steam is currently offering two games free to keep, but the promotion ends on January 22. Once claimed, both titles remain in the user’s library permanently, even if they are never downloaded.

The two games available at no cost are Crown Champion: Legends of the Arena and Battle Simulator: Counter Stickman. Users only need to add them to their Steam library before the deadline to keep them.

Crown Champion: Legends of the Arena is a hybrid RPG and management game centered on gladiatorial combat. Players recruit and train fighters, manage resources, and progress through a series of arenas while building reputation. The game was originally released in 2016 and normally sells for a low price, but it has maintained a largely positive reception on Steam.

Battle Simulator: Counter Stickman takes a simpler approach. Players command small groups of stickman units in short battle scenarios. The game is relatively brief and has mixed user reviews, but the limited time commitment makes it easy to try without risk. As with many free-to-keep promotions on Steam, trading cards are included, but card drops are disabled.

To keep either game, users must manually claim it on Steam before January 22. There is no requirement to install them immediately. Once added, the titles stay tied to the account and can be downloaded at any time in the future.

Steam frequently runs free-to-play promotions, especially outside major sales periods. Users who track these offers often add games to their libraries even if they do not plan to play them right away, since there is no downside once the claim is made.

Users who want both games should act before the promotion ends, as Steam typically removes the free option without extensions once the deadline passes.

Thank you for being a Ghacks reader. The post Steam Users Can Claim Two Free Games Until January 22 appeared first on gHacks Technology News.

Evernote has released Evernote 11, the first major version update of the note-taking application in roughly five years. The release bundles a large number of background changes made over the past two years and adds three new AI-driven features that affect how notes are searched, summarized, and created.

Evernote 11 keeps the app’s core structure intact. Notes, tasks, and organizational tools work as before, but new AI components are layered on top. The update is positioned as a functional shift rather than a redesign of how users store information.

One of the main additions is the AI Assistant. The feature appears as a chat-style panel inside Evernote and allows users to ask questions about their existing content. It can summarize long notes, surface related entries, and pull together information across notebooks and tasks. The assistant also supports basic web searches from within the app. Evernote says the feature was developed in collaboration with OpenAI.

Another change affects search behavior. Semantic Search replaces strict keyword matching with natural language queries. Instead of remembering exact terms used in a note, users can describe what they are looking for and receive results based on meaning rather than phrasing. This applies across notebooks and works with longer or loosely defined queries.

The third AI feature focuses on meetings. AI Meeting Notes can record and transcribe conversations from in-person or online meetings, detect multiple speakers, and generate summaries that include key points and action items. The resulting output is saved directly as a note, removing the need to move transcripts between apps or services.

Each AI feature can be enabled or disabled individually through new settings. AI Assistant is available on desktop and web. Semantic Search and AI Meeting Notes are supported on desktop, web, and mobile platforms, according to Evernote.

Outside of AI, Evernote 11 includes a set of general changes to editing, navigation, and content capture. Evernote has not listed all of these individually but describes them as refinements based on the steady stream of updates shipped during the development of version 11. Syncing, performance, and reliability improvements from those updates are now considered part of the new baseline.

The update also introduces a refreshed visual design. Evernote has adjusted parts of the interface and rolled out a new logo. The layout remains familiar, but visual elements have been modernized to align with the new release.

Evernote says work on version 11 began around two years ago. Rather than shipping a single large update earlier, the company continued to release incremental improvements and folded them into this version alongside the new AI features.

Evernote 11 is available now for download on supported platforms. Users who do not want AI features enabled can turn them off, but the new search and assistant tools are part of the default installation.

Do you still use Evernote for note-taking?

Thank you for being a Ghacks reader. The post Evernote 11 Introduces AI Assistant, Semantic Search, And Meeting Notes appeared first on gHacks Technology News.

Microsoft has issued a series of emergency out-of-band updates to address serious problems introduced by the January 2026 Patch Tuesday updates. The affected patches caused shutdown and hibernation failures on some systems and disrupted access to remote desktop-based cloud services.

The issues surfaced shortly after Microsoft released cumulative updates on January 13, 2026. Users reported erratic behavior on both consumer and enterprise systems, prompting Microsoft to acknowledge the problems and publish fixes outside the normal update cadence.

One of the confirmed issues affects remote access to Azure Virtual Desktop and Windows 365. After installing the January updates, some systems were no longer able to establish remote desktop connections, preventing users from accessing cloud-hosted Windows environments.

A second issue impacts devices with Secure Launch enabled. On these systems, shutdown and hibernation no longer worked as expected. Instead of powering down, affected machines would restart repeatedly, making it difficult to fully shut down the system.

Due to the scope of the problems, Microsoft released multiple standalone updates rather than waiting for the next Patch Tuesday. The company published six separate out-of-band updates covering Windows 11 versions 23H2 through 25H2, Windows 10 version 22H2, and supported Windows Server releases from 2019 through 2025.

The emergency updates are currently available through the Microsoft Update Catalog. Microsoft has indicated that the fixes will be included in the regular February 2026 cumulative updates, reducing the need for separate installations later.

Additional issues were reported alongside the confirmed bugs, including Outlook failing to start, black screens during login, and other stability problems. Microsoft has not confirmed whether these behaviors are directly related to the January updates or addressed by the out-of-band fixes.

Systems affected by shutdown failures or broken remote desktop connections are expected to return to normal behavior after installing the emergency updates. Administrators managing enterprise environments may need to deploy the fixes manually if automatic updates are deferred or controlled through update policies.

Thank you for being a Ghacks reader. The post Microsoft Releases Emergency Updates After Breaking Core Features appeared first on gHacks Technology News.

OpenAI has changed how ChatGPT uses past conversations. Plus and Pro subscribers can now ask questions that reference their entire chat history, including conversations created months or years ago, without manually opening older threads.

ChatGPT has stored chat history for a long time, but finding specific information inside it has been unreliable. Searching manually often returned unrelated threads, especially when conversations covered similar topics. The new system improves how ChatGPT retrieves and uses past context when answering questions.

The update allows ChatGPT to actively pull relevant details from previous conversations when responding. Users can ask follow-up questions like continuing a recipe, revisiting a workout plan, or resuming a technical setup discussed earlier, and ChatGPT can now locate that information on its own.

When past conversations are used, ChatGPT displays them as clickable references. These links open the original chat, allowing users to review the full context or confirm where a specific detail came from. This makes it easier to verify answers and revisit earlier discussions without guessing which thread contained the information.

The feature is currently limited to ChatGPT Plus and Pro subscriptions. Free users still have access to chat history, but without the enhanced retrieval and referencing behavior.

OpenAI refers to the underlying system internally as PersonalContextAgentTool. The goal is to move away from treating each chat as isolated and instead allow ChatGPT to behave more like a continuous assistant that remembers prior interactions when relevant.

The update brings ChatGPT closer to parity with Google Gemini, which added past conversation referencing in early 2025. Gemini users have been able to continue long-running conversations with historical context for some time, making this a notable catch-up move for ChatGPT.

OpenAI has been rolling out broader changes alongside this feature, including newer models and interface updates. The history search change directly affects daily usage by reducing friction when returning to older topics and removing the need to manually organize or bookmark important chats.

For users who rely on ChatGPT as a long-term reference tool rather than a one-off chatbot, the change alters how conversations can be reused. Past discussions now function more like a searchable knowledge base tied to the account, rather than a static archive that requires manual navigation.

How often do you use ChatGPT? Every day, or on a weekly basis?

Thank you for being a Ghacks reader. The post ChatGPT Now Searches Your Full Chat History For Answers appeared first on gHacks Technology News.

Manually copying files from SD cards is a weak point in many backup routines. Cards get reused before files are copied, folders are created inconsistently, or footage stays on a laptop longer than intended. A NAS-based workflow can remove those steps by copying files automatically when an SD card is inserted.

The goal is simple: insert an SD card, let the NAS handle the copy, and remove the card once the process finishes. No file managers, no drag-and-drop, and no guessing whether everything was transferred.

Some NAS devices make this easy by design. Systems with built-in SD or microSD card slots can be configured to copy new files to a fixed location with a button press or automatically on insertion. Once the copy completes, the card can be ejected safely. Files land directly on network storage and are immediately available to other devices on the network.

This setup works best when the NAS is already the main storage location for photos or video. Instead of copying files to a laptop first and moving them later, footage goes straight to centralized storage. That reduces duplication and lowers the chance of working on files that only exist on removable media.

Most NAS systems do not include SD card readers, but the same workflow is still possible. A USB SD card reader plugged into the NAS can be treated like any other removable storage device.

On Synology systems, the USB Copy package handles this task. Once installed, it can be configured to watch for newly connected USB devices and automatically copy their contents to a predefined folder. The process can run without user interaction, which makes it suitable for repeated imports from cameras or recorders.

QNAP offers a similar feature through Hybrid Backup Sync. A backup job can be set up to trigger when a USB device is connected. Files can be copied or moved to internal storage based on the chosen policy. As with Synology, the card reader itself is generic; the automation happens entirely on the NAS.

Other NAS platforms often provide comparable plugins or background services. The naming and setup steps vary, but the concept is consistent: detect removable storage, copy files, and store them in a predictable location.

If the NAS does not support automated USB copy at all, the workflow can shift to a connected computer. This still removes manual copying, even though the automation runs on the desktop instead of the NAS.

On macOS, file monitoring tools can watch for mounted volumes with specific names. When an SD card appears, files are copied automatically to a mounted network share on the NAS. This works best when SD cards use consistent volume labels, which allows the automation to trigger reliably.

Windows and cross-platform setups can achieve the same result using synchronization tools that support folder monitoring. A batch or sync job is created first, then a companion service watches for the SD card mount and runs the job when detected. Once configured, the process runs in the background without further input.

A basic automated workflow usually follows these steps:

1. Assign a consistent name or label to SD cards.
2. Mount the NAS as a network drive or share.
3. Configure a monitoring tool to watch for the SD card mount.
4. Define a copy or sync job to a fixed NAS folder.
5. Remove the card after the transfer completes.

This approach takes more time to set up initially, but it behaves the same way as built-in NAS automation once configured.

Automating SD card backups reduces reliance on memory and habit. Files are copied the same way every time, cards can be reused sooner, and data is protected by the NAS as soon as it leaves the camera. Users who work with multiple cards or move between different machines benefit the most, especially when the NAS is already part of the editing or archiving workflow.

Thank you for being a Ghacks reader. The post Automate SD Card Backups To A NAS And Skip Manual File Copying appeared first on gHacks Technology News.

Microsoft has issued emergency out-of-band updates for Microsoft Windows 11 after the latest security patches introduced two serious regressions. Affected systems were unable to shut down or hibernate correctly, and Remote Desktop sign-ins failed on certain configurations.

The issues appeared with Microsoft’s first Windows 11 updates released in 2026. Reports quickly surfaced from users who found that shutdown attempts would hang or fail entirely. Separately, Remote Desktop connections could no longer authenticate, preventing sign-ins to affected machines. Both problems were severe enough that Microsoft released fixes outside of the normal update schedule.

Microsoft has now published two out-of-band updates, KB5077744 and KB5077797, which are rolling out through Windows Update. Installing the updates restores normal shutdown behavior and resolves the Remote Desktop authentication failures.

The shutdown issue affects devices running Windows 11 version 23H2 with Secure Launch enabled. On these systems, shutdown and hibernation could fail after the January security updates were installed. The Remote Desktop problem has a broader scope, affecting Windows 11 version 25H2, Windows 10 22H2 ESU, and Windows Server 2025, according to Microsoft’s advisory.

Users who rely on Remote Desktop to access work or home systems were effectively locked out until the emergency patches became available. Microsoft has not provided a detailed explanation for how the issues passed testing, but confirmed that the fixes fully address the reported behavior.

The out-of-band updates are offered automatically via Windows Update. Devices configured to pause updates or managed through enterprise update policies may require manual approval or deployment before the fixes are applied.

These issues add to a growing list of Windows 11 update problems seen over the past year. Previous updates have caused Task Manager to remain running after being closed, leading to resource usage problems over time. Another update introduced a bright white flash when opening File Explorer in dark mode, despite being marketed as an improvement to dark mode behavior.

Microsoft has also had to fix an update that broke the Windows Recovery Environment, preventing users from accessing recovery options when troubleshooting system failures. Each of these issues was addressed after release, but only once users encountered them on production systems.

For now, Microsoft recommends that affected users install the emergency updates as soon as they become available. Systems that cannot shut down properly or accept Remote Desktop connections are expected to return to normal behavior immediately after the patches are applied.

Thank you for being a Ghacks reader. The post Emergency Windows 11 Updates Fix Shutdown And Remote Desktop Failures appeared first on gHacks Technology News.

Mozilla has released Mozilla Firefox 147, making the update available ahead of its official announcement scheduled for January 13, 2026. The release focuses on display rendering fixes on Linux, media playback behavior, and a collection of platform-specific improvements that affect performance, privacy, and compatibility.

Linux users running GNOME with Mutter will notice one of the most visible changes. Firefox 147 now matches its window and rendering surface sizes to the actual pixel grid used by the compositor. On fractionally scaled displays, this removes the slight blur that could appear at scaling values such as 125% or 150%. The improvement applies regardless of window size and does not require manual configuration.

Another Linux-related change is support for the Freedesktop.org XDG Base Directory Specification. Firefox now follows standard directory locations more closely, which affects where certain data and configuration files are stored. Users with custom scripts or backups tied to specific paths may want to double-check them after updating.

Media playback receives attention in this release as well. Firefox 147 enables zero-copy hardware-decoded video on AMD GPUs. By reducing unnecessary memory transfers during playback, the browser can lower CPU usage and improve performance, especially when streaming high-resolution video.

Firefox’s phishing and malware protection has also been updated. The browser now supports the Safe Browsing V5 protocol, replacing older versions used to retrieve and process threat data. The change happens in the background and does not alter user-facing controls, but it modernizes how Firefox handles malicious site detection.

On Apple Silicon Macs, Firefox 147 enables WebGPU support by default. This allows websites and web applications to access more advanced GPU features for rendering and compute tasks without relying on experimental settings. The feature is limited to compatible hardware and supported content.

Page loading performance can improve on supported sites thanks to the addition of Compression Dictionaries, based on IETF RFC 9842. This technology allows shared dictionaries to be reused when compressing content, reducing the number of bytes transferred. The benefit is most noticeable on slower or bandwidth-constrained connections.

Firefox 147 also changes how Picture-in-Picture works. When a video is playing in a tab and that tab is sent to the background, Firefox now opens a Picture-in-Picture window automatically. The feature was previously hidden behind Firefox Labs and is now enabled for all users. Those who prefer manual control can adjust Picture-in-Picture behavior in settings.

The browser’s settings layout has been adjusted slightly. Tab-related options are now grouped into Opening, Interaction, and Closing categories. The reorganization does not change existing functionality but makes related options easier to find.

Profile management becomes more accessible in Firefox 147. A new Profiles section appears under General settings, allowing users to create and manage separate profiles for different use cases. Each profile keeps its own bookmarks, passwords, and browsing data, which can help separate work and personal activity or support multiple users on the same system.

With one experimental feature graduating to stable, Firefox Labs gains a replacement. A new “Lists and timer on Firefox Home” experiment adds simple productivity tools to the start page, such as timers and reminders. The feature remains optional and disabled by default.

Several smaller fixes and behavior changes are included. Firefox 147 resolves an issue that caused some HTTP/3 requests with non-UTF-8 headers to time out. Draggable buttons now behave correctly when dragging starts directly on the button, and Accept-Language headers support the same quality values used by other browsers.

Privacy behavior changes for users running Enhanced Tracking Protection in Strict mode. Local network access restrictions are now enabled by default, requiring explicit permission before public websites can reach local network resources. This may affect access to local devices or self-hosted services through the browser.

On Android, Firefox 147 enables Site Isolation by default, aligning mobile protection against certain side-channel attacks with desktop Firefox. On Windows, the update improves compatibility with some monitors that previously caused tab selection problems.

Developer tools also receive incremental updates. View Transitions now expose related pseudo-elements and animations more clearly, the JSON viewer can send resources directly to Firefox Profiler, and CSS tooling improves support for pseudo-elements, anchor positioning, and position-try fallbacks. Firefox 147 also expands standards support, including new CSS features, Navigation API support, module service workers, Unicode 17, and additional locales.

Mozilla plans to formally announce Firefox 147 alongside new ESR releases on January 13, 2026. Installation packages for supported platforms are already available, and users affected by the Linux rendering changes or updated privacy defaults may want to review settings after upgrading.

Thank you for being a Ghacks reader. The post New Firefox 147 Is Now Available For Download appeared first on gHacks Technology News.

Apple’s iPad lineup enters 2026 in a strong position. Over the past year, Apple refreshed every iPad model with newer chips, meaning buyers can now choose between tablets powered by A-series, M-series, and even Apple’s latest silicon, depending on their needs and budget.

The current lineup includes four main models: the standard iPad, iPad mini, iPad Air, and iPad Pro. While all of them run iPadOS 26 and support Apple’s modern software features, they target very different use cases.

Below is a breakdown of the best iPads to buy in 2026 and who each model is best suited for.

Best iPad for Most Users: iPad Air (M3)

The iPad Air with Apple’s M3 chip is the most balanced option in Apple’s tablet lineup. It offers performance close to the iPad Pro while staying significantly more affordable.

The M3 chip provides more than enough power for everyday tasks, creative work, multitasking, and even demanding apps
like photo and video editors. The tablet is available in both 11-inch and 13-inch sizes, includes a landscape front
camera, and supports Apple Pencil Pro and the latest Magic Keyboard.

For users who want a tablet that can double as a lightweight laptop without paying Pro prices, the iPad Air is the
safest recommendation.

Best iPad for Professionals: iPad Pro (M5)

The iPad Pro with Apple’s M5 chip is aimed squarely at professionals and power users. It is the fastest and most capable iPad Apple has released to date, with performance that rivals many laptops.

Key highlights include the XDR Retina display with HDR support, optional nano-texture glass for reduced glare, and support for high-end creative workflows such as video editing, 3D work, and advanced illustration. Storage options go up to 2TB.

While the iPad Pro offers unmatched performance, it is also the most expensive model in the lineup. It makes sense primarily for users who know they will benefit from the extra power and display quality.

Best Budget iPad: iPad (11th generation)

The standard iPad remains Apple’s most affordable tablet and continues to offer strong value in 2026. While it lacks the raw performance of the M-series models, it is more than capable for streaming, browsing, light gaming, schoolwork, and everyday use.

It includes a USB-C port, solid cameras, and compatibility with Apple accessories like the Magic Keyboard Folio. Its lower price also makes it a popular choice for families and children.

For users who want an iPad at the lowest possible cost and do not need professional features, the base iPad is still an excellent option.

Best iPad for Reading and Portability: iPad mini (7th generation)

The iPad mini remains the most compact and portable iPad available. Its smaller size makes it especially well-suited for reading, note-taking, and one-handed use.

Despite its size, the iPad mini is not underpowered. It supports Apple Intelligence features and offers solid performance for media consumption and casual tasks. However, its smaller display makes it less suitable as a laptop replacement, and it does not support Apple’s Magic Keyboard.

Users who prioritize portability and reading comfort over multitasking should consider the iPad mini.

Which iPad Should You Buy?

Choosing the right iPad depends largely on how you plan to use it:

• Most users: iPad Air (M3)
• Professionals and creators: iPad Pro (M5)
• Budget-conscious buyers and families: iPad (11th generation)
• Readers and travelers: iPad mini (7th generation)

For most people, the iPad Air strikes the best balance between price, performance, and versatility. The other models make sense when specific needs—such as professional workloads, portability, or budget—take priority.

Thank you for being a Ghacks reader. The post The Best iPads to Buy in 2026 appeared first on gHacks Technology News.

Microsoft has released the first security updates of 2026 for its Windows operating system and other Microsoft products. The January Patch Tuesday addresses more than 110 security vulnerabilities across Windows, Microsoft Office, Edge, Azure components, and additional services.

The updates are available for all supported Windows versions. Windows 10 continues to receive security fixes only under Microsoft’s Extended Security Updates (ESU) program, while Windows 11 versions 24H2 and 25H2 are the primary consumer releases receiving updates.

Our monthly Patch Day overview provides home users and system administrators with a concise summary of released updates, vulnerability severity, known issues, and direct download resources.

Microsoft Security Updates — January 2026 Executive Summary:

• Microsoft released approximately 112–114 security updates in total
• Around 8 vulnerabilities are rated Critical
• At least one zero-day vulnerability is actively exploited, confirmed by Microsoft
• Security fixes cover Windows, Office, Edge, Azure, and server components

Windows clients with known issues

• Windows 11: versions 23H2, 24H2, 25H2

Windows Server products with known issues

• Windows Server 2023
• Windows Server 2025

Product Overview

Below is a breakdown of affected Windows versions and reported vulnerability counts.

Windows 11:

Windows 11 version 24H2

• Dozens of vulnerabilities fixed
• Includes remote code execution, privilege escalation, spoofing, and information disclosure flaws

Windows 11 version 25H2

• Same vulnerability scope as 24H2
• No critical vulnerabilities specific to this release alone

Windows 11 version 23H2 (Enterprise & Education)

• Still supported through enterprise servicing channels
• Receives cumulative security updates only

Windows 10

Windows 10 (Extended Security Updates only)

• Applies only to systems enrolled in ESU
• No feature updates
• General support ended in October 2025

Windows Server

• Windows Server 2016
• Windows Server 2019
• Windows Server 2022
• Windows Server 2023
• Windows Server 2025

All supported server versions received January security updates. No critical vulnerabilities were disclosed publicly for server products, but multiple important vulnerabilities were addressed.

Windows Security Updates Windows 11 version 24H2 and 25H2

Support page: KB5074109

Updates and improvements:

• Security fixes for multiple remote code execution and elevation of privilege vulnerabilities
• Improvements to system stability and reliability
• Servicing changes related to Secure Boot and virtualization-based security

Windows 11 version 23H2 (Enterprise & Education)

Support page: KB5073455

Updates and improvements:

• Security fixes and quality improvements
• Includes changes related to Secure Boot certificates

Known issue:

• Systems with Secure Launch enabled may fail to shut down or enter sleep mode
• Devices may restart instead

Workaround:

Use a command prompt and run:

shutdown /s /t 0

Microsoft has no workaround for sleep mode at this time.

Windows 10 (ESU)

Support page: KB5073724

• Security-only update
• Available exclusively to ESU-enrolled systems
• Removes outdated modem drivers as part of servicing

Windows Server Updates

• Windows Server 2025: KB5073379
• Windows Server 2023: KB5073450

Security updates only, no feature changes.

Servicing Stack and Quality Updates

• KB5077800 — Servicing Stack Update (released January 17, 2026)
  • Improves update installation reliability for older Windows builds

Microsoft Office Security Updates

Microsoft released Office and server component security updates on January 13, 2026.

Office updates include:

• Excel 2016 — KB5002831
• Office 2016 — KB5002826
• Word 2016 — KB5002829

SharePoint Server:

• SharePoint Server Subscription Edition — KB5002822
• SharePoint Server 2019 — KB5002825
• SharePoint Server 2016 — KB5002828

Language packs received corresponding updates.

Known Issues After Installing January Updates Remote Desktop / Azure Virtual Desktop / Windows 365

After installing the January updates, some users report:

• Credential prompt failures
• Authentication issues in Azure Virtual Desktop and Windows 365
• Problems primarily affect the new Windows App

Workarounds:

• Use the classic Remote Desktop client
• Use the web-based RDP client

Microsoft has acknowledged the issue and is actively investigating.

Windows 11 23H2 Shutdown and Sleep Issues

• Affects devices with Secure Launch enabled
• Prevents normal shutdown and sleep behavior
• Documented in Windows Release Health

Microsoft has released out-of-band fixes for some affected systems as of January 18, 2026.

How to Install the January 2026 Updates Immediately

Security updates are rolled out automatically, but manual installation is possible.

1. Open the Start menu
2. Type Windows Update and open it
3. Select Check for updates
4. Choose Download & install all

Recommendation:

Create a full system backup before installing updates, especially on production systems.

Direct Update Downloads Windows 11

• KB5074109 — Cumulative Update for Windows 11 24H2 / 25H2
• KB5073455 — Windows 11 23H2 (Enterprise & Education)

Windows 10 (ESU)

• KB5073724 — January 2026 ESU update

Additional Resources

• December 2025 Security Updates
• Microsoft Security Update Guide
• Windows 11 24H2 Update History
• Windows 11 25H2 Update History
• In-depth Windows Update installation guide

Thank you for being a Ghacks reader. The post The Last Windows Security Updates of January 2026 Are Here appeared first on gHacks Technology News.

After years of resisting a feature that rivals have offered for a long time, Google Chrome is finally getting vertical tabs. The feature, which allows tabs to be displayed along the side of the browser instead of across the top, has now surfaced in Chrome Beta, signaling that a broader rollout is close.

Chrome has long been the odd one out among major browsers when it comes to tab management. Browsers such as Microsoft Edge, Firefox, Brave, and others have supported vertical tabs for years, making it easier to manage large numbers of open tabs, especially on widescreen displays. Despite Chrome's massive market share, Google has been slow to adopt the layout.

That appears to be changing.

From Canary to Beta

Google was first spotted experimenting with vertical tabs in Chrome Canary back in late 2025. At the time, the feature was clearly experimental and limited to early testers. Now, it has made its way into Chrome Beta (version 145), suggesting it is nearing readiness for general users.

The feature is still hidden behind an experimental flag, but its presence in the Beta channel usually means Google is confident enough to begin wider testing.

How vertical tabs work in Chrome

Once enabled, Chrome's entire tab strip moves from the top of the window to the side. All open tabs appear in a vertical list, which can make it significantly easier to scan titles, especially when many tabs are open.

Users can switch back to the traditional layout at any time. A simple right-click in the tab area lets you toggle between vertical tabs and the classic horizontal strip, so the change is fully optional.

How to try vertical tabs right now

If you're using Chrome Beta or Canary, you can already test the feature:

1. Open chrome: //flags/#vertical-tabs
2. Enable the Vertical Tabs flag
3. Restart the browser
4. Right-click the tab bar and select Show tabs on the side

The option does not yet appear in the stable version of Chrome.

When it comes to stable Chrome

Based on Google's regular release schedule, vertical tabs are expected to arrive in the stable version of Chrome around January 28. Assuming no major issues arise during Beta testing, this will mark the first official release of vertical tabs to Chrome's massive user base.

Why this matters

For users who juggle dozens of tabs, vertical tabs can dramatically improve usability. They make better use of horizontal screen space, reduce tab shrinking, and often pair well with tab grouping and pinned tabs.

Chrome finally adopting this feature brings it closer in line with competitors and removes one of the more common complaints from power users.

If all goes as planned, one of Chrome's most requested interface features will soon be available to everyone-better late than never.

Thank you for being a Ghacks reader. The post Google Chrome is Finally Embracing Vertical Tabs appeared first on gHacks Technology News.