AI News
2.2K views | +40 today
Your new post is loading...
Your new post is loading...
Scooped by Philippe Gassmann
August 12, 8:37 PM
Scoop.it!

General Catalyst leads $1.1B round into 2-month-old River AI

General Catalyst leads $1.1B round into 2-month-old River AI | AI News | Scoop.it
River AI, a startup founded by xAI co-founder Igor Babuschkin, has a fascinating vision for personal agents and secured $1.1 billion out of the gate.
No comment yet.
Scooped by Philippe Gassmann
August 12, 7:37 AM
Scoop.it!

[2608.10430] Actionable Hallucination Detection: Translating Latent Uncertainty into Agentic Critique

[2608.10430] Actionable Hallucination Detection: Translating Latent Uncertainty into Agentic Critique | AI News | Scoop.it
Abstract page for arXiv paper 2608.10430: Actionable Hallucination Detection: Translating Latent Uncertainty into Agentic Critique...
No comment yet.
Scooped by Philippe Gassmann
August 12, 7:37 AM
Scoop.it!

[2608.10366] DSAgentBench: Can Agents Automate End-to-End Data-Science Workflows in Real Computer Environments?

[2608.10366] DSAgentBench: Can Agents Automate End-to-End Data-Science Workflows in Real Computer Environments? | AI News | Scoop.it
Abstract page for arXiv paper 2608.10366: DSAgentBench: Can Agents Automate End-to-End Data-Science Workflows in Real Computer Environments?
No comment yet.
Scooped by Philippe Gassmann
August 11, 7:07 AM
Scoop.it!

Lucrative AI Launches Out of Stealth with $500K in Pre-Seed Funding to Bring MCP-Native Automation to Enterprise Revenue Operations

Lucrative AI Launches Out of Stealth with $500K in Pre-Seed Funding to Bring MCP-Native Automation to Enterprise Revenue Operations | AI News | Scoop.it
Lucrative AI Inc. has launched its enterprise revenue engine, backed by a $500,000 incubation investment from Mountainise Inc.The platform uses Model Context...
No comment yet.
Scooped by Philippe Gassmann
August 11, 7:07 AM
Scoop.it!

elvis on X: "I respect this a lot. There isn't a lot of focus on agent quality, review, and observability. @NuphosAI looks like a clean AI-native DevOps workspace, and I like that it's built as a c...

I respect this a lot. There isn't a lot of focus on agent quality, review, and observability. @NuphosAI looks like a clean AI-native DevOps workspace, and I like that it's built as a collaborative tool for agents and engineers.
No comment yet.
Scooped by Philippe Gassmann
August 11, 7:07 AM
Scoop.it!

Nous Research on X: "Hermes has twelve browser tools. Browser Use mode replaces them with a single one, driven by @browser_use's CLI 3.0.
Instead of a dozen schemas in every request and a tool cal...

Hermes has twelve browser tools. Browser Use mode replaces them with a single one, driven by @browser_use's CLI 3.0.

Instead of a dozen schemas in every request and a tool call per click, the agent writes a script.
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

PRELOOP CI_ENGINE

>_ PRELOOP Drop-in, agent-native GitHub Actions that runs locally, or self-hosted. Verify code at the speed agents write it. Preloop is an agent-native CI engine, built in Rust. Every job runs in its own hardware-isolated microVM that boots in under 200 milliseconds on macOS, Linux, or Windows. The same engine runs locally or self-hosted: a workflow that passes on your machine passes in CI. INSTALL_SCRIPT COPY_AGENT_INSTRUCTIONS JOB: #8492_AGENT_PR TIME_ELAPSED: 1.16s 00:00.00 SYS REQ_RECEIVED PR #402 00:00.01 PROV ALLOCATING MICRO_VM (linux-arm64) 00:00.17 OK VM_BOOT_COMPLETE [174ms] 00:00.18 EXEC PULLING REPO /agent-generated-patch 00:00.24 EXEC RUN: cargo test --workspace 00:01.01 OUT test result: ok. 142 passed; 0 failed; 00:01.07 EXEC RUN: npm run lint 00:01.14 OUT All files pass linting. 00:01.15 OK WORKFLOW_COMPLETE [1.15s] 00:01.16 SYS TEARDOWN_INITIATED 01 THE_BOTTLENECK THE OLD VERIFICATION STACK WAS BUILT FOR HUMAN-SPEED OUTPUT. Legacy CI was built for human-speed output: minutes to provision a runner, containers that share your host kernel, and failure pages designed for manual inspection. When an AI agent ships code all day, every change is untrusted by default. Waiting four minutes for a runner to boot nullifies the advantage of automated generation. 02 EXECUTION_MODEL [A] GHA_PARITY { .YML } → || Drop-in GitHub Actions compatibility. Your existing workflows run unchanged, verified against the official runner, not approximated. [B] MICROVM_ENV [ μVM ] Every job gets its own dedicated microVM. A real kernel, hardware-isolated, not a shared one. Boots in under 200 ms on macOS, Linux, or Windows. [C] LIVE_DEBUG >_ ↔ █ Pause-on-failure. When a job fails, the machine freezes. Open a shell at the exact failure state, or fork the machine and retry from the step that broke. [D] DAP_DEBUG ▸_ ↔ DAP Built-in Debug Adapter Protocol bridge. Attach a DAP client like VS Code to a paused job and step through the failure. Built in, coming soon. [E] X86_ON_ARM x86 → arm64 Run x86_64 workflows on an Apple Silicon Mac. Rosetta translation keeps them near native speed, so amd64 CI does not need an amd64 machine. [F] WARM_STATE [ ▓▓▓░ ] After the first run the whole filesystem stays warm. Dependencies, compiled output, and package caches are still there, so the next run skips the install phase entirely. 03 ARCHITECTURE_DELTA LOCAL-FIRST PARITY The same engine runs on your laptop and on self-hosted workers. If it passes locally, it passes in CI. Zero environment drift. Hand a warm machine snapshot to a bigger host and continue with every cache intact. FULL STACK OWNERSHIP The runner, the control plane, and the CLI are built in Rust. One fast binary, no Node runtime, no JVM, no orchestration sprawl. AGENT-OPERABLE API Built to be operated by agents, not just triggered. DENY-BY-DEFAULT Network is off until you switch it on. Secrets are referenced from the host at launch, so they are never baked into a machine or its snapshots. Agent-generated code stays quarantined. 04 THE_AGENT_LOOP RUN WHAT IS ON YOUR DISK. preloop run tests your working tree as it is, uncommitted and untracked files included. Checkout inside the job pulls your local tree instead of a pushed commit, and your git index is never touched. THE FAILED JOB STAYS ALIVE. A failing step holds its microVM open by default, with the checkout, toolchain, services, and warm caches exactly as the test left them. There is nothing to reproduce. FIX IT, RETRY THE STEP. preloop debug attaches to the paused step. Edit the code, run :retry, and only that step runs again. Build caches survive, the debris from the failed attempt does not. TAKE THE FIX WITH YOU. Edits you made inside the VM come back to your workspace as a patch with preloop debug --export. Add --json and an agent drives the identical loop. PRELOOP CLI LOOP: RUN → FIX → VERIFY $ preloop run → VERDICT: FAILED, step 3/8 (cargo test) → VM PAUSED, caches warm, services up $ preloop debug › :retry → RE-RAN STEP 3/8 ONLY → VERDICT: PASS [1.2s] $ preloop run --sync → PUSHED TESTED SHA, PR #402 UPDATED 05 NO_WEBHOOK_REQUIRED THE SHA THAT LANDS ON GITHUB IS THE SHA THAT WAS TESTED. Every other CI waits for GitHub to hand it a webhook. When that delivery path breaks, a tunnel drops, a secret rotates, or GitHub has an incident, CI goes dark and the missed events never come back. Preloop inverts it. You submit the run straight to the server and it starts immediately. GitHub becomes the place results land, not the thing runs wait on. PUSH_CONFIRM. preloop run --sync runs CI on your workspace, then pushes the tested commit and confirms it on the PR with check runs. The push happens after the verdict, not before. ONLY TESTED CODE GETS PUSHED. When the run finishes, the CLI pushes the exact commit it tested. Fast-forward or a new branch only. A force push never happens on its own. THE PR UPDATES ITSELF. The server opens or updates the pull request, draft by default, and reports check runs against the tested commit. SAFE TO REPLAY. GitHub unreachable? The push retries on its own, and preloop sync replays the whole thing. Every step is idempotent. 06 SYSTEM_SPECS GHA_COMPATIBILITY Drop-in for existing workflows. Property-tested, formally verified, with wire compatibility steps. COLD_START_LATENCY < 200 ms to a running microVM PLATFORMS macOS (Apple Silicon) · Linux · Windows WORKSPACE Runs your working tree, uncommitted and untracked files included CONTROL_PLANE Rust end to end: runner, control plane, CLI ISOLATION One dedicated microVM per job; network off by default MEM_FOOTPRINT Elastic. The host commits only what a job actually uses SELF_HOSTING preloop serve runs the control plane and microVM pool on your own box 07 DEMOS [A] DROP_IN An existing GitHub Actions workflow, run as-is. No config changes, no rewrites. [B] DEBUG_LOOP A failing job, a split terminal. Fix the file on the host, rerun, and the same warm machine goes green. [C] AGENT_LOOP An agent drives the loop: submit, read the verdict, patch, resubmit. Machine-readable end to end. 08 ACT_VS_AKSH ACT REIMPLEMENTS THE RUNNER. WE REIMPLEMENTED THE CONTROL PLANE AND THE RUNNER, WITH THE REAL WIRE PROTOCOL BETWEEN THEM. act is a pragmatic Go binary that runs steps in Docker containers on your host kernel. Fast to start, but it cannot guarantee parity: no wire protocol, no OIDC, no concurrency groups, no debugger. On the same 39-scenario benchmark, act gets 29/39 right, aksh gets 31/39 — with typical runs in 400-800 ms. FEATURE ACT AKSH SCOPE Runner only (Go CLI) Control plane + runner (Rust) FIDELITY Behavioral, mostly works Official wire protocol, the runner cannot tell the difference EXECUTION Docker containers on your kernel One dedicated microVM per job VERIFICATION Community bug reports Golden wire captures replayed, differential + property tests BENCHMARK 29/39 scenarios (74%) 31/39 scenarios (79.5%), sub-second typical runs DEBUGGING None DAP debugger, breakpoints, stepping OIDC + CONCURRENCY Not supported Full OIDC provider, concurrency groups PLATFORM Linux Docker only macOS, Linux, Windows FULL COMPARISON → DOCS/ACT_VS_AKSH.MD (act vs agent-ci vs aksh vs Gitea Actions, with the full 39-scenario benchmark) INSTALL_CLI curl -fsSL https://preloop.dev/install.sh | sh [ GITHUB ] © PRELOOP_SYS 2026
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

Moore Threads Turns a 147 Percent Half Into a Hong…

Moore Threads Turns a 147 Percent Half Into a Hong… | AI News | Scoop.it
Moore Threads posted 147.42 percent first half revenue growth and a near breakeven result before filing to issue H shares in Hong Kong.What it means for…...
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

[2608.06714] The Optimizer Is the Agent: Reasoning-Driven Search across Prompts, Programs, and ML Workflows

[2608.06714] The Optimizer Is the Agent: Reasoning-Driven Search across Prompts, Programs, and ML Workflows | AI News | Scoop.it
Abstract page for arXiv paper 2608.06714: The Optimizer Is the Agent: Reasoning-Driven Search across Prompts, Programs, and ML Workflows...
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

[2608.07169] Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory

[2608.07169] Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory | AI News | Scoop.it
Abstract page for arXiv paper 2608.07169: Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory...
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

[2608.07430] Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits

[2608.07430] Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits | AI News | Scoop.it
Abstract page for arXiv paper 2608.07430: Diffusion LLMs as Targets and Adversaries: Mechanistic Safety Exploits...
No comment yet.
Scooped by Philippe Gassmann
August 12, 7:37 AM
Scoop.it!

[2608.10875] VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World?

[2608.10875] VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World? | AI News | Scoop.it
Abstract page for arXiv paper 2608.10875: VibeLifeBench: Can Your Life Agent Be Proactive and Persistent in a Living World?
No comment yet.
Scooped by Philippe Gassmann
August 12, 7:37 AM
Scoop.it!

[2608.11047] V-FiLLM: Verified Financial LLM Reasoning Benchmark

[2608.11047] V-FiLLM: Verified Financial LLM Reasoning Benchmark | AI News | Scoop.it
Abstract page for arXiv paper 2608.11047: V-FiLLM: Verified Financial LLM Reasoning Benchmark...
No comment yet.
Scooped by Philippe Gassmann
August 12, 7:37 AM
Scoop.it!

[2608.11095] Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding

[2608.11095] Why Does CLAUDE.md Keep Growing? Catastrophic Remembering in Agentic Coding | AI News | Scoop.it
Abstract page for arXiv paper 2608.11095: Why Does CLAUDE.md Keep Growing?Catastrophic Remembering in Agentic Coding...
No comment yet.
Scooped by Philippe Gassmann
August 11, 7:07 AM
Scoop.it!

NAVER D2SF Makes Follow-On Investment in NdotLight, a Physical AI Data Startup

NAVER D2SF Makes Follow-On Investment in NdotLight, a Physical AI Data Startup | AI News | Scoop.it
-NdotLight builds data infrastructure that can be directly used for physical AI training simulations...
No comment yet.
Scooped by Philippe Gassmann
August 11, 7:07 AM
Scoop.it!

From Coding to Finance: Microsoft’s AI Agent Push

Microsoft says AI agents are already fundamentally changing how its software engineers work — and sales, finance and other knowledge workers are next. Charles Lamanna, Microsoft EVP of Copilot, Agents and Platform, joins Bloomberg to explain how employees are shifting from doing tasks themselves to overseeing AI agents, why the company sees the technology driving more productivity rather than simply reducing headcount, and what’s behind Microsoft 365 Copilot’s surge past 30 million paid seats. He joins Ed Ludlow on "Bloomberg Tech."
--------
Like this video? Subscribe to Bloomberg Technology on YouTube:
https://www.youtube.com/channel/UCrM7B7SL_g1edFOnmj-SDKg
 
Watch the latest full episodes of "Bloomberg Technology" with Caroline Hyde and Ed Ludlow here:
https://www.youtube.com/playlist?list=PLfAX25ZLrPGTygCwn55voYZ_LYyKjxokJ
 
Get the latest in tech from Silicon Valley and around the world here:
https://www.bloomberg.com/technology
Connect with us on...
X: https://twitter.com/technology
Facebook: https://www.facebook.com/BloombergTechnology
Instagram: https://www.instagram.com/bloombergbusiness/
 
Follow Ed Ludlow on X here: https://twitter.com/EdLudlow
Follow Caroline Hyde on X here: https://twitter.com/CarolineHydeTV
 
Listen to the daily Bloomberg Technology podcast here:
https://www.bloomberg.com/podcasts/series/bloomberg-technology
 
More from Bloomberg Business
Connect with us on...
X: https://twitter.com/business
Facebook: https://www.facebook.com/bloombergbusiness
Instagram: https://www.instagram.com/bloombergbusiness/
LinkedIn: https://www.linkedin.com/company/bloomberg-news/
TikTok: https://www.tiktok.com/@bloombergbusiness
No comment yet.
Scooped by Philippe Gassmann
August 11, 7:07 AM
Scoop.it!

Anthropic on X: "We asked an unreleased research version of Claude to take a stab at the Riemann hypothesis.
It didn’t solve it, but it did make strides on a related problem: it increased the lowe...

We asked an unreleased research version of Claude to take a stab at the Riemann hypothesis.It didn’t solve it, but it did make strides on a related problem: it increased the lower bound for the fraction of zeros of the Riemann zeta function that satisfy the hypothesis from...
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

Gotcha | Samosa-AI

Gotcha | Samosa-AI | AI News | Scoop.it
Gotcha is a fully on-device AI copilot for Android — control your phone with natural language through 100+ device tools, dual safety modes, and voice calls over any app.
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

Simon Willison on X: "@_cartermp OpenAI's models had to find two separate zero-days in Artifactory to break free, so more than just misconfiguration in that case" / X

@_cartermp OpenAI's models had to find two separate zero-days in Artifactory to break free, so more than just misconfiguration in that case...
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

[2608.07468] SimWAM: A Simple World Action Model for End-to-End Autonomous Driving

[2608.07468] SimWAM: A Simple World Action Model for End-to-End Autonomous Driving | AI News | Scoop.it
Abstract page for arXiv paper 2608.07468: SimWAM: A Simple World Action Model for End-to-End Autonomous Driving...
No comment yet.
Scooped by Philippe Gassmann
August 10, 9:09 AM
Scoop.it!

[2608.07077] Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking

[2608.07077] Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking | AI News | Scoop.it
Abstract page for arXiv paper 2608.07077: Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking...
No comment yet.