Most AI stops at answers. You AI assistant Hiro delivers decision-ready artifacts. From source-backed analysis to visual insights, Hiro transforms your research into HTML pages, presentation slides, and full reports your team can review and share.Less time formatting. More time making decisions. Watch how Hiro turns evidence into something your team can actually use #AIAgents# #Research# #EngineeringAI# #PatsnapEureka#

Future-built companies are already engineering AI value. BCG’s 2026 Applied AI Index breaks down what separates these companies into a formula with three core elements: strategic clarity plus applied AI can translate AI ambition into transformative impact. These future-built companies scale five times as many AI workflows to full value creation as laggards do. Discover the details in the full report:

I crunched 16,600 X posts and 514 podcasts to write The State of AI Report. It covers the 10 trends shaping AI adoption going into 2027: 01 Horizontal Agents - Grok Bot and the merging of the model, harness, and application layers 02 Knowledge Work Factories - building loops so your agents can achieve your goals while you sleep 03 The Verification Gap - turning taste into a test 04 Open Models - cost, specialization, and privacy fuel demand 05 The Context Layer - your company's playbooks and expertise become infrastructure 06 Generative Media - advancements in voice, video, and image realism 07 AI-Native Services - sell the outcome not the output, build agents to scale 08 Vibes Shift - from Dario fueled replacement fear to building the factory 09 Bubble Talk - what funding and capex mean for vendor selection 10 Deployment - the rise of forward deployed engineering AI didn't take summer vaca All layers of the stack improved. And so did best practices for AI adoption inside companies. Read this practical report that separates AI signal from noise, shows where AI creates value today, and helps prioritize investments for 2027:

AI just proved that the Exponential Time Hypothesis (ETH) is false. Cryptography relies on ETH, but luckily this particular discovery doesn't threaten core algorithms yet. However, ETH was thought true for almost 30 years, and is highly relevant in the field. This illustrates how AI is the true threat to crypto and security, because it can move the ground under our feet, and it's here today. Currently, many chains are promoting their migration to "quantum safe" signatures and other schemes. Meanwhile, cryptographically-relevant quantum computers won't be around for some years, and even when they are, they'll be specialized machines, requiring special environments, rather than easily accessible mass market devices, and it will be relatively easy for authorities and regulatory frameworks to control their use, at least initially. Renting them for the purposes of hacking won't be allowed, for obvious reasons (not that this is something we should rely upon, but it impacts any objective threat analysis). However, there are dangers with adopting these schemes. The less well-studied and more complex a cryptography algorithm, the greater the chance that a sudden shift in our understanding of theoretical computer science or mathematics, or sophisticated analysis by AI, will suddenly reveal a catastrophic bug. The first real-world harbinger of the dangers occurred when AI found issues with HAWK, a quantum-safe algorithm that was passing through NIST's standardization process at the time (it was withdrawn). Chain operators' desire to ride narratives and exploit the tendency of web3 investors tendency to seize onto this or that carefully promoted story, who for the most part do not use crypto technology beyond holding and trading tokens, and who lack understanding of the theoretical issues involved, is certainly exploitative for a simple reason — whereas historical technical safety manias used for commercial purposes, such as Y2K, only collected money, this one also risks user safety. Chains sell adoption of quantum safe signature schemes to token buyers and users that in reality reflect a kind of stupidity, and irresponsibility, rather than actions reflecting wise stewardship protecting people from an imminent quantum meltdown. Meanwhile, of course, while adding support for a new post quantum signature algorithms doesn't involve much engineering work, it allows them to post pictures of their logo next to photos of quantum computers, so they can sell the ignorant on how sci-fi and advanced they are, which makes it particularly attractive to them. Important note: you'll hear about some serious people performing post quantum encryption of messages. That's different. They're calculating the risk that adversaries, such as state adversaries, perform mass storage of encrypted messages containing secrets, so that they can extract the secrets from them 10 years from now, say, when they finally have cryptographically-relevant quantum computers. This is a different kind of risk they are addressing. The perverse tragedy of the quantum mania craze that's currently gripping crypto is that on the one hand it shows how so many chains are driven by a desire to sell tokens first, rather than pursue sound engineering practices, and on the other, their willingness to lie and risk user safety in the process. This, all while the real wolf is at the door as concerns our engineering — AI that's rapidly advancing, here today, accessible to many, and difficult to control. What I can promise, is that DFINITY will be focusing on mitigating this real risk, and no matter the post quantum mania, we won't be joining it, and will instead try to plot the most responsible course.

Engineering, Product, Marketing, each in its own silo, with its own context OpenClaw Multiplayer gives teams one shared agent and shared context, so they can actually work together @Pishgar on what we showed at GitHub Universe:

Engineering leaders told Grepr founder and CEO Jad Naous they spend 5 to 20% of their infrastructure budget on observability, and it's buying less and less reliability. Jad, a former a16z partner, joins a16z's Martin Casado to discuss whether AI can run production systems as well as it writes code. Grepr sits in a company's log stream, can scan a petabyte of logs a day, cuts observability costs by about 90%, and is building AI agents that use those signals to troubleshoot production issues. How AI is changing engineering at Jad's 10-person company: - A data lake feature that once took one engineer about six months took Jad 10 days - The gap between okay and great engineers on his team "just multiplied 10X" - He runs 3-4 agent sessions at once, with a "tech lead" agent reviewing every PR Why running live systems is a harder problem for AI: - Jad: AI can almost tune databases, but can't yet keep servers running on its own - Martin: every bit of code doubles the ways a program can behave, which is why humans still own systems - The real world is messier than AI's training data, self-driving took 20 years and $170B 0:55 From Stanford to a16z to Grepr 7:00 Why observability spend stopped paying off 11:00 From autocomplete to AI that trains engineers 13:10 Why AI widens the engineer skill gap 14:55 Syntax vs semantics 18:00 Inside Jad's AI workflow 22:55 Six months of work in 10 days 24:50 Can AI reason about how systems really run? 30:10 Why AI SRE tools start in the wrong place 32:30 Can AI run a distributed cluster yet? 37:35 Why engineers buy extra capacity over predicting 39:15 How AI could trace failures in production 43:45 Why AI for ops is harder than for code 49:55 Can AI do a VC's job? 52:05 Why self-driving took 20 years and $170B 57:05 Customers skipped the AI pitch for the 90% savings YouTube: @jadtnaous @grepr_ai @martin_casado

Engineering T cells for immunotherapy vs cancer had already had marked success in certain types. But there's a next generation in the hopper, as reviewed today @Cancer_Cell, that ought to help take the field much further into solid tumors