Former Google CEO Eric Schmidt and his wife, Wendy, have announced major investments in four telescope projects. Each of the telescopes brings a new novel capability online. The most intriguing instrument is a space-based telescope named Lazuli that is designed to be a more capable and modern version of the Hubble Space Telescope. The Schmidts' investment, estimated to be worth half a billion dollars, has enormous potential to advance the study of astronomy and astrophysics.
TVs today are vastly cheaper than they were 25 years ago, even before adjusting for inflation. The price per area-pixel has fallen by more than 90%. It's not surprising that the cost of the technology declined. Virtually all of the major mechanisms that can drive efficiency improvements have occurred.
This post shares predictions Simon Willison made on a recent podcast. The predictions cover what he thinks will happen in the next 1, 3, and 6 years in the tech industry. There has never been so much uncertainty about what's coming in the next year. The significant advances in coding agents in just the last two months indicate that things will change significantly, but it is unclear what those changes will be.
Grok Build is a coding agent from xAI that can be accessed both on the web and through a CLI. It currently supports local agents, with remote environment support marked as coming soon. There are options for users to configure agents and custom environments. A video previewing the feature is available in the article.
China is reviewing Meta's recent acquisition of AI startup Manus, a warning to Chinese entrepreneurs interested in Silicon Valley riches. The review is part of broader consideration by China on how it can build fences around its AI know-how. The government is identifying standout companies and technologies with the aim of adding those of strategic importance to an export-control list. Manus' parent company was set up by Chinese co-founders in China, but it moved its headquarters to Singapore last year to expand its global user base.
AWS isn't dying - it isn't even struggling in any conventional sense. However, its era of unchallenged dominance is over. The company is navigating a transition that would challenge any organization. It has made the right strategic moves, but its success will depend on whether it can retain the people who know how to deliver.
Elon Musk's lawsuit alleges that OpenAI's leadership made knowingly false assurances to Elon Musk about its charitable mission that they never honored in favor of their personal self-enrichment.
Type systems have become a common safety feature for developers working with AI because static types ensure that code is more trustworthy and more maintainable.
If AI code reviewers are supposed to save time, why am I triaging 40 comments about trailing whitespace, "consider using an f-string," and "line exceeds 100 characters" on a 10-line change?
Gmail is adding AI Overviews using Gemini technology, summarizing email threads and allowing users to query their inbox with natural language. Its new features, Help Me Write, Suggested Replies, and Proofread, improve email drafting, while AI Inbox filters prioritize key messages. These updates, available for free, are rolling out now in the US.
OpenAI is acquiring the team behind Convogo, a business software platform that automates and improves leadership assessments and feedback reporting. The startup will not acquire Convogo's IP or technology. Convogo's team will join OpenAI to work on its AI cloud efforts. Its product will be wound down.
The poster for 'Lord of War', which depicts an image of Nicholas Cage comprised of bullets, is a masterclass in compositing. A team of designers likely had to manually source images of bullets, match the lighting of the individual bullets with the shadows on the face, and arrange thousands of shell casings to form a recognizable human face. That level of skill was rare in 2005. This article explores whether AI can redesign a similar 'Lord of Tokens' image with GPU dies and silicon wafers constituting Jensen Huang's face.
This post demonstrates how to implement long-term memory and stateful personalization in AI agents using OpenAI's Agents SDK. Structured memory objects and context-injection patterns enable agents to recall preferences, consolidate user notes, and evolve over time.
Increased efficiency in resource use can lead to higher total consumption, but only when demand is elastic. Cheaper production just changes who captures the surplus when demand is constrained by something other than production cost. AI coding will change how things are built without changing how much gets built or at what quality for a huge amount of economically significant software. This post looks at eight markets to see what impact AI coding will have.
Why is AI stuck in pilot mode? It's constrained by data infrastructure and enterprise context. CData surveyed 200+ data & AI experts [10] on what's working (and breaking) in AI data connectivity. The key finding? IF YOUR DATA ISN'T LIVE, CONTEXTUAL, AND GOVERNED, YOUR AI WON'T DELIVER. Read the 2026 State of AI Data
Warp specialization is a popular technique that improves kernel performance on GPUs out of the box. The key idea is to have specialized code paths for each warp instead of the same code. This reduces performance hits due to control flow divergence, improves latency hiding, and makes better use of hardware units on the GPU. This post outlines the current design of warp specialization in Triton, a compiler that aims to generate performance-portable code and runtime across hardware for AI kernels.
This guide walks readers through the process of building a code review agent from scratch that can read files, run commands, edit code, and figure out the steps to accomplish a task. The resulting agent can analyze a code base, find bugs and security issues, and return structured feedback. The guide aims to help readers understand how the Claude Code SDK works so they can build whatever they actually need.
Agentic Rubrics enable test-free verification of code changes by creating checklist-style evaluations from expert agents that interact with codebases. They outperform baseline methods on SWE-Bench Verified and offer a scalable, interpretable signal for reinforcement learning and inference-time feedback.
Modern AI systems have evolved past simply predicting the next word and now demonstrate enhanced capabilities, such as solving complex problems and reasoning. Systems like OpenAI and Google DeepMind have achieved impressive feats, including outperforming top mathematicians in competitions, reflecting significant advancements in cognitive tasks beyond autocomplete.
Coding agents like Gemini 3 Pro and Claude Opus 4.5 have automated tasks ranging from invoice creation to predicting US corn yields. These agents exhibit capabilities close to artificial general intelligence but require human understanding and oversight to avoid errors and maximize their potential. The rise of these tools signals a shift towards enabling individuals to independently explore complex software projects. They demonstrate the importance of foundational technical knowledge and creating opportunities and challenges for both new and incumbent firms.
Developers are shipping more code - with PRs getting bigger and denser. Read Greptile's cross-industry study [17] on the state of AI coding to see AI tool adoption, model growth trends, and model snapshots.
The entire Electric Coin Company development team resigned after a governance clash with the Bootstrap board and said they are forming a new company to continue building “unstoppable private money.” Zcash's roadmap and organizational future are now quite uncertain, even though the protocol itself is said to be unaffected.
Wyoming launched FRNT on Solana, the first US state-backed stablecoin managed by Franklin Templeton, held in a Wyoming trust exclusively in USD and short-term Treasuries. Interest from reserves funds Wyoming's school program as the token bridges to six EVM chains, including Ethereum, Arbitrum, Base, Optimism, Polygon, and Avalanche via LayerZero. The launch follows the 2023 Stable Token Act and an 11-blockchain consultation period concluding with Solana's selection as the native chain.
Aptos launched confidential transactions on devnet through a Distributed Lab partnership, using homomorphic encryption and zero-knowledge proofs to hide transaction amounts while maintaining regulatory compliance via designated auditor keys that enable independent decryption access. The system encrypts transfer amounts under three separate keys and allows validators to verify ZK proofs and update balances without seeing underlying values, allowing for institutional use cases like private payroll, position-masked trading, and non-signaling treasury operations.
Fixed-rate demand is driven by borrowers (institutions and loopers) who want predictability, not by lenders, and on-chain lenders prize withdrawal-anytime flexibility, so locking capital for term loans forces unsustainably high fixed yields and kills scale. The scalable solution is a rate-exchange layer that leaves Aave/Morpho/Euler liquidity intact while letting traders hedge and synthesize fixed payments (high capital efficiency, margin-backed exposure), enabling borrowers to buy certainty without forcing lenders into term locks. If implemented, rate-exchanges would unlock institutional credit expansion on-chain by matching fixed-rate demand to rate traders rather than fixed-rate depositors.
There are four privacy infrastructure shifts for 2026: privacy chains will establish durable competitive moats because secrets cannot bridge between chains without metadata leakage risks, creating network effects unavailable to commodity L1s competing solely on performance. Decentralized messaging protocols require quantum-resistant encryption combined with censorship-resistant architecture to survive state-level shutdowns, giving users cryptographic ownership of communications. Confidential data handling for finance and healthcare is moving from application-level solutions to core infrastructure through programmable access rules, client-side encryption, and decentralized key management. DeFi security is evolving from reactive bug-hunting to formal specification enforcement, where AI-assisted proof tools establish global invariants that become runtime guardrails automatically reverting transactions that violate safety properties.
People are frustrated with legacy banks. There's a demand for a single stablecoin neobank that truly replaces banks, has a physical card, enables ACH pull with named bank accounts, Solana deposits/withdrawals, has the ability to send wires/ACH (multi-currency), and ideally noncustodial.
Reputation systems, ICO renaissance, and crypto neobanking are poised to have a strong 2026. Reputation infrastructure will emerge as the critical filter for distinguishing humans from AI bots, with projects like Ethos, Legion, and Kaito building social-capital-based systems that require hundreds of hours and genuine relationships to establish credibility. ICOs are resurging after 2025's regulatory clarity resolved the failures of airdrop farming and low-float VC models, with reputation-gated mechanisms preventing the scam-filled dynamics of 2017. Crypto neobanking will compete directly with TradFi by offering self-custodial accounts, yield-bearing opportunities exceeding traditional 0.2% APR savings rates, and instant cross-border stablecoin payments as stablecoin adoption accelerates.
Banks pay near-zero on savings, T-bill yields fall with rate cuts, and stocks are unpredictable, so DeFi, where assets are being tokenized, is the place to earn meaningful yield and generate permissionless passive income through lending, borrowing, and yield strategies.
Ethereum's validator entry queue has climbed above 1.45M ETH with a ~25-day wait, while exits remain minimal, implying staking demand is roughly 1,800× larger than withdrawals.
The new left navigation bar in Figma Design, Draw, and Dev Mode improves navigation between library assets, variables, and search with toggleable labels available in the View menu. A redesigned variable authoring experience now offers an edge-to-edge full-screen view that spans the entire browser width.
Google is rolling out a new AI Inbox for Gmail that surfaces suggested to-dos and grouped updates, adds natural-language AI Overviews to inbox search, and introduces a Grammarly-like Proofread tool, while keeping the traditional inbox optional. Many AI features previously limited to paid tiers are expanding to all users, with advanced tools like AI search overviews and Proofread initially available to Google AI Pro and Ultra subscribers.
OpenAI is reportedly testing advertisements within ChatGPT using internal employee-only chat models. The company has been exploring ad layouts and sponsored response disclosures despite CEO Sam Altman telling employees in December that public-facing app advertising plans were on hold. OpenAI aims to create a new type of digital advertising that maintains user trust while providing stable funding for the company.
AI tools now enable non-designers to create visual designs that appear 90% complete, commoditizing what was once a designer's exclusive superpower: visualization. However, these AI-generated designs still require the critical 10% of iterative refinement that only trained designers can provide by understanding and articulating the "why" behind design decisions. Designers' evolving role centers on bridging user needs, business goals, and technical constraints while justifying design choices that AI cannot yet evaluate or explain.
MTCHMKR is a consumer SaaS platform that connects brands for creative partnerships that addresses the disjointed process across Slack communities. It is a polished product with subscription management, AI-assisted matchmaking, and mobile functionality. The team launched a beta at $1/month to overcome the cold-start problem, gathering waitlist users through social media. It has planned future AI agents for deeper brand research and partnership strategy.
UX research isn't becoming obsolete in the age of AI - it's shifting upstream to define what meaningful, high-quality AI outputs look like by grounding prompts, formats, and evaluation criteria in real user needs and context. Researchers play a strategic role in shaping, guiding, and holding LLM-based systems accountable for delivering genuine value by translating human expectations into clear quality signals and rubrics.
Wellows is a single source of truth for brand visibility across AI search engines. It helps teams understand how AI interprets their presence by unifying citations, sentiment, and visibility data into one clear platform.