Tech Roundup: Google I/O 2024, Twitter Becomes X.com & More

[A recurring feature on the latest in Science & Technology.]

• OpenAI rolls out AI updates to GPT-4 as it seeks to get ahead of Google, intensifying competition among companies for an edge in emerging technology; announces GPT-4o (o for omni), a new flagship generative AI model that is faster and natively multimodal (i.e., integrates text, vision and audio capabilities) for free to all ChatGPT users, opens its GPT Store to all users for free, including the ability to create custom GPTs, and debuts a dedicated app for Apple macOS as well as makes it possible for paying users to directly upload files from Google Drive and Microsoft OneDrive.
• Microsoft's carbon emissions grows 29.1% from 2020 to 2023 amid its push to be the global leader in AI, imperiling its goal to be carbon negative by 2030.
• Apple releases iOS 17.5 with cross-platform tracking device detection and the ability for users in the E.U. to download apps directly from websites; also brings an offline mode to Apple News+ and includes a repair state option that lets an ‌iPhone‌ be sent in for service without deactivating Find My and Activation Lock. (It's worth noting that Apple has removed the Apple News widget in regions where the services is officially not available.)
• Google partners with HP to ship 3D video conferencing service Project Starline in 2025 with plans to support Zoom and Meet; adds a new featre to Google Drive called hovercards that allows users to preview information about files without opening them.
• Microsoft unveils Places, an AI-powered app to help employees on Outlook and Teams better coordinate in-office time; updates paste option in Word for Windows to apply "Merge Formatting" by default.
• Anthropic launches its Claude AI chatbot and subscription plans in Europe for both individuals and businesses.
• The European Commission opens a probe into Meta over concerns that it may have "breached the Digital Services Act (DSA) in areas linked to the protection of minors" and that the company's Facebook and Instagram products "may stimulate behavioural addictions in children, as well as create so-called 'rabbit-hole effects'"; designates Booking as a gatekeeper under the Digital Markets Act (DMA), saying "holidaymakers will start benefiting from more choice and hotels will have more business opportunities."
• Consumers protection groups across the E.U. file coordinated complaints against Temu, accusing the Chinese e-commerce platform of a raft of breaches related to the DSA, including rules against manipulative design, and transparency around product recommender algorithms.
• X officially begins redirecting users from twitter.com to x.com; rolls out a revamped version of its Communities feature, which lets users network around topics of interest, alongside improved discovery tools and recommendations.
• Smartphone shipments in India grow 8% YoY in Q1 2024; Vivo leads for the first time with 19.2% market share, followed by Xiaomi with 18.8% and Samsung with 17.5%.
• ByteDance TikTok tests a more robust search results page with AI generated results for select queries; tests 60-minute videos, which would let media companies post full TV show episodes and put TikTok in even more direct competition with YouTube.
• X (formerly Twitter) rival Bluesky pilots a new "Show more like this" and "Show less like this" tool in the Discover feed to help users fine-tune their algorithmic timeline.
• Microsoft faces a legal setback after a U.S. court says the company must pay IPA Technologies US$ 242 million after finding that the Cortana virtual assistant software infringed on a patent related to voice recognition that expired in January 2019; tech giant says it will appeal the decision.
• Meta-owned Instagram expands its creator marketplace to 10 new countries, including South Korea, Germany, Netherlands, France, Spain, Israel, Turkey, Mexico, Argentina and Indonesia; to shut down its enterprise tool Workplace at the star of September 2025, making it read-only until May 2026, before it's completely decommissioned, and recommends organisations to migrate to Zoom-owned Workvivo.
• Google experiments with running ChromeOS on Android within a virtual machine; open-sources code for Project Gameface, its hands-free gaming "mouse" controlled by head movement and facial gestures.
• Google overhauls search at Google I/O to display AI-generated responses to queries by default and rolls out a search filter, named "Web," to show only text-based links in results; announces a chatty new voice mode with Gemini Live, and voice and video search with Google; updates Circle to Search to solve physics and math problems; and unveils deeper AI integration in Android with scam call detection and answer questions about videos and PDFs being viewed on screen.

• Google unveils Music AI Sandbox, a suite of music AI tools that include the ability to create short audio clips or "stems" based on text prompts; debuts next-generation text-to-image model Imagen 3, a new generative AI video model called Veo that "can generate 'high-quality' 1080p resolution videos over a minute in length in a wide variety of visual and cinematic styles," and an upgraded SynthID watermark imprinting system to identify digitally generated text and video.
• Google brings its Gemini AI model to Calendar, Chrome, Docs, Gmail, Google Photos, Keep and Tasks, and teases Project Astra, a camera-based AI app to process visual data in real time; announces plans to allow Gemini Advanced subscribers to make personalised Gemini chatbots, called Gems, with distinct personalities and abilities.
• Google launches Android 14 for TV, with new energy modes, accessibility features including color correction and picture-in-picture for some TV models; unveils new Android device theft protection features, options to add items to Google Wallet from a photo, and an AI-powered live threat detection mechanism to flag malware.
• Google opens up API access to its Google Home smart home platform; releases Project IDX, its next-gen AI-centric browser-based IDE.
• Reddit launches new tools for its Ask Me Anything Q&A feature, including a dedicated AMA tab, support for up to five guest hosts, and promotional options, and brings back its awards system after shutting down in July 2023; inks deal with OpenAI to bring Reddit content to ChatGPT and more via Reddit's Data API.
• Apple says it prevented over US$ 7 billion in "potentially fraudulent transactions" on the App Store from 2020 through 2023, counting more than US$ 1.8 billion in 2023, down from US$ 2 billion in 2022.
• Google adds a new feature to Sheets that lets users "quickly format and organise data with tables"; to formally discontinue Google One's VPN service on June 20, 2024. (Pixel 8 and newer devices offer a built-in VPN, similar to Apple iOS' Private Relay. A similar functionality is expected to be available in early June for Pixel 7, 7 Pro, 7a and Fold devices.)
• Movie streamer Netflix announces plans to launch an in-house advertising technology platform, as its ad-supported tier surpasses 40 million users globally.
• Broadcom, which recently acquired VMware, makes VMware Fusion Pro 13 and VMWare Workstation Pro free for personal use; says "users who want a virtual lab on their Mac, Windows or Linux computer can do so for free simply by registering and downloading the latest build."
• The National Music Publishers Association (NMPA) sends a cease and desist letter to Spotify, accusing the music streaming service of using its members' copyrighted content without appropriate licensing.
• Chipmaker Intel debuts Thunderbolt Share, a new specification and app to connect two PCs, letting users transfer data at 40Gbps+, control another PC and sync files.
• Apple previews accessibility features, including Eye Tracking, which lets users with physical disabilities control iPads or iPhones with their eyes, and Vehicle Motion Cues, which adds animated dots to iPhone and iPad screens to reduce motion sickness for passengers using them in moving vehicles.; announces improvements to the Magnifier app with Reader Mode and showcases Music Haptics that "plays taps, textures and refined vibrations to the audio of the music" to assist users who are hard of hearing to experience music on iPhone.
• Meta rolls out a fact-checking program for Threads, allowing users to rate and mark false content on the platform; faces criticism for facilitating a thriving black market by allowing users to buy and sell Facebook accounts on the social media network that can be used to run political ads in India, in violation of its policies, per an investigation from the Tech Transparency Project.
• Payment processor Visa rolls out tech that will allow it to share more data about customers' preferences based on their shopping history with retailers via proprietary "tokens" (it's worth noting that the sharing requires the consumer's consent); adds a passkey service to help confirm a user's identity during online transactions using biometrics.
• Neuralink's recent disclosure that a "number of threads retracted from the brain" of its first patient is an issue the brain implant company has known about for years, according to a report from Reuters.
• Lenovo and Motorola get hit with a temporary sales ban in Germany amid ongoing dispute with InterDigital over a patent infringement regarding WWAN modules.
• Meta starts testing TweetDeck-style customisable feeds on Threads that are stacked in a column interface on the web for selected users; also pilots an Instagram feature called Peek that lets users post unedited, authentic pictures that can only be viewed once.
• Google launches a Gemini add-on for educational institutes via Workspace that offers integration with Docs, AI-powered note-taking and data loss prevention; bring new accessibility features to Lookout to help users find objects and capture images, debuts a new Look to Speak app for Android for communicating with images and symbols, and updates Maps to surface wheelchair-accessible entrances and find places that can cast to hearing devices.
• Media player Winamp announces plans to open source the application's source code on September 24, 2024.
• Google's DeepMind AI division releases its Frontier Safety Framework, a set of protocols for "proactively identifying future AI capabilities that could cause severe harm and putting in place mechanisms to detect and mitigate them"; comes as OpenAI dissolves its Superalignment team, which was announced in July 2023 to focus on long-term AI risks, leading to senior executive departures over concerns that "safety culture and processes have taken a backseat to shiny products" at the company.
• Apple limits the development and testing of third-party browser engines to devices physically located in the E.U., forcing browser makers to have staff in the bloc.
• Salesforce-owned Slack sparks privacy outcry after it confirms it is training some of its AI-powered features, but not its generative AI tool, on user content and uploads, with all users opted-in by default.
• Terrorists associated with the Islamic State are believed to have leveraged AI to broadcast extremist propaganda quickly and cheaply in the form of video dispatches called News Harvest that's hosted on an unnamed private platform, according to a report from the SITE Intelligence Group.