I Gave Gemini Spark Access to My Life. Google's new AI agent combed through my emails, documents, and calendar to plan a birthday party and still didn't clock the person most important to me. At its recent I/O developer conference, Google introduced Gemini Spark as an always-on agent that connects to your personal data, completes online tasks, and automates aspects of your daily interactions. It's Google's take on the viral OpenClaw agent that rocked Silicon Valley at the start of 2026. OpenClaw's early adopters handed their entire lives over to an AI agent for messaging and scheduling automation--sometimes with bot-induced mishaps causing embarrassing results.

Customer data from more than 350 hotels around the world may have been accessed as part of realistic reservation-hijacking scams. Travelers' information and booking details may have been stolen from hundreds of hotels around the world, according to new findings from security researchers. These swiped trip details, such as booking names and reservation information, are then being repurposed by cybercriminals to create highly targeted phishing messages used to steal credit card information. At least 350 hotels, vacation rentals, motels, and guesthouses in 50 different countries have been caught up in so-called reservation hijacking scams, according to an analysis of phishing messages and cybercriminal infrastructure by security company Norton. Researchers say the use of legitimate booking information in phishing messages may increase the chances that someone clicks on a fraudulent link and hands over other sensitive details to criminals.

Don't forget the bear spray. More information Adding us as a Preferred Source in Google by using this link indicates that you would like to see more of our content in Google News results. Breakthroughs, discoveries, and DIY tips sent six days a week. The park, spread across 2.2 million acres and three states, includes half of Earth's active geysers, the Grand Canyon of the Yellowstone River, and stunning wildlife. Ahead of the 2026 summer tourist season, Yellowstone National Park recommends following these 10 steps for making the most out of your visit.

The US Built a Site to Ensure Fair Access to Public Lands. Recreation.gov was supposed to make access to public lands more equitable and streamlined. It's a few minutes before 8 am Mountain Time on March 16, the day that river permit cancellations are released on Recreation.gov, the federal website for public land reservations. Rec.gov, as it's commonly called, administers everything from river permits and timed entrance fees at the most popular national parks to campground reservations on remote sites belonging to the Bureau of Land Management, and a lot of people are recreating on public land these days. There were 11 million reservations on the site in 2024, up significantly from 3.5 million reservations reported in 2019. At the center of it all is an unlikely player in the outdoor recreation space: The site is operated by the government contractor Booz Allen Hamilton, a corporation known more for cybersecurity than rafting trips. Early each year, outdoor enthusiasts gear up for Recreation.gov's annual lotteries for some of the most iconic experiences in the country: a river trip down Idaho's Middle Fork of the Salmon River, which flows through the Frank Church River of No Return Wilderness. Backcountry permits to hike into the Wave, an otherworldly rock formation in Arizona's Paria Canyon-Vermilion Cliffs Wilderness. Overnight stays in the rugged, lake-studded Enchantments, in Washington's Okanogan-Wenatchee National Forest. Odds of getting a desirable Middle Fork permit are around 2 percent.

When you purchase through links in our articles, we may earn a small commission. I asked AI to book dinner. ChatGPT, Claude, and Gemini may be aces at coding, but they're less than magical when it comes to booking a table for three. I can clearly see the day when we'll be able to summon ChatGPT, Claude, or Gemini on our phones, say something like "Hey ChatGPT, book a table for two at Outback Steakhouse tonight at 8," and ChatGPT will simply take care of it. All of the big AI providers are busy unveiling integrations for everyday services ranging from Spotify and DoorDash to AllTrails and the dinner reservation app Resy, with varying degrees of success.

After the rise (and fall) of reservation scalping, DoorDash and a host of apps are fighting to book you a seat at the country's most exclusive restaurants. At The Eighty-Six in Manhattan, exclusivity is the point. The luxe, 11-table steakhouse is the sort of place that lavishes caviar and aged mimolette cheese on its potatoes, and crows that your market-price duck was raised by one Dr. Taylor Swift has reportedly dined there in a Miu Miu skirt. Reservations are a scarce commodity that the restaurant, and New York law forbids you from selling one. "Access is the main asset," wrote food writer Helen Rosner in a recent New Yorker review of The Eighty-Six. "The product is the door, and what a door!

Top-10 rejected images in the MNIST testing set found by two methods.

This was supposed to be the year when autonomous agents took over everyday tasks. One year ago, Sam Altman, the C.E.O. of OpenAI, made a bold prediction: "We believe that, in 2025, we may see the first AI agents'join the workforce' and materially change the output of companies." A couple of weeks later, the company's chief product officer, Kevin Weil, said at the World Economic Forum conference at Davos in January, "I think 2025 is the year that we go from ChatGPT being this super smart thing . . . to ChatGPT doing things in the real world for you." He gave examples of artificial intelligence filling out online forms and booking restaurant reservations. He later promised, "We're going to be able to do that, no question."

Despite rapid progress in LLM agents, performance on long-horizon, tool-using tasks remains fragile. To better understand this fragility, we ask a simple question: \emph{do all actions contribute equally to failure?} Analyzing execution traces on $τ$-Bench (Airline/Retail) and SWE-Bench Verified, we decompose trajectories into \emph{mutating} (environment-changing) vs.\ non-mutating steps and formalize \emph{decisive deviations}, earliest action, level divergences that flip success to failure. A logistic regression reveals that each additional deviation in a mutating action reduces the odds of success by upto $92\%$ on Airline and upto $96\%$ on Retail for SoTA models. In contrast, deviations in non-mutating actions have little to no effect. Errors also grow with context length as agents drift from role and act on stale constraints. Motivated by these observations, we introduce \cm{}, a model-agnostic, gradient-free, test-time safeguard that (i) adds mutation-gated verification, (ii) injects \emph{Targeted Reflection} before mutating steps, and (iii) performs block-based context cleaning. \cm{} delivers consistent gains, e.g., Qwen3-Thinking: +28\% \emph{relative} on Airline, +11\% on Retail, and +7\% on SWE-Bench Verified; Claude: +9\%/+7\%. We further identify ceiling effects in $τ$-Bench, where annotation errors and underspecified tasks artificially cap model performance. To address this, we release $τ$-Bench Verified, which restores benchmark headroom through targeted revisions. Our results argue for action-level analysis, targeted safeguards, and reliable evaluations as prerequisites for robust multi-turn agents.

Tool agents interact with users through multi-turn dialogues to accomplish various tasks. Recent studies have adopted user simulation methods to develop these agents in multi-turn settings. However, existing user simulators tend to be agent-friendly, exhibiting only cooperative behaviors, which fails to train and test agents against non-collaborative users in the real world. To address this, we propose a novel user simulator architecture that simulates four categories of non-collaborative behaviors: requesting unavailable services, digressing into tangential conversations, expressing impatience, and providing incomplete utterances. Our user simulator can simulate challenging and natural non-collaborative behaviors while reliably delivering all intents and information necessary to accomplish the task. Our experiments on MultiWOZ and $τ$-bench reveal significant performance degradation in state-of-the-art tool agents when encountering non-collaborative users. We provide detailed analyses of agents' weaknesses under each non-collaborative condition, such as escalated hallucinations and dialogue breakdowns. Ultimately, we contribute an easily extensible user simulation framework to help the research community develop tool agents and preemptively diagnose them under challenging real-world conditions within their own services.