Author: Naveed Ahmad

If in case you have ever stared at 1000’s of strains of integration take a look at logs questioning which of the sixteen log recordsdata truly comprises your bug, you aren’t alone — and Google now has information to show it. A workforce of Google researchers launched Auto-Diagnose, an LLM-powered software that robotically reads the failure logs from a damaged integration take a look at, finds the basis trigger, and posts a concise analysis immediately into the code evaluation the place the failure confirmed up. On a handbook analysis of 71 real-world failures spanning 39 distinct groups, the software accurately…

Jack Zhang was 34 years outdated, three and a half years into working a startup, and sitting throughout from one of the crucial highly effective traders in Silicon Valley. Michael Moritz of Sequoia had invited him to his residence — a spot with, Zhang remembers, a few flooring and a view straight to the Golden Gate Bridge — to make the case for promoting. Stripe needed to purchase Airwallex for $1.2 billion. On the time, the Melbourne firm had round $2 million in annualized income. The maths was nearly fairly irresistable: a income a number of someplace close to 600…

Hackers have damaged into not less than one group utilizing Home windows vulnerabilities printed on-line by a disgruntled safety researcher over the past two weeks, in keeping with a cybersecurity agency. On Friday, cybersecurity firm Huntress stated in a series of posts on X that its researchers have seen hackers making the most of three Home windows safety flaws, dubbed BlueHammer, UnDefend, and RedSun.  It’s unclear who the goal of this assault is, and who the hackers are. BlueHammer is the one bug among the many three vulnerabilities being exploited that Microsoft has patched thus far. A repair for BlueHammer…

On this tutorial, we discover the best way to run OpenAI’s open-weight GPT-OSS fashions in Google Colab with a powerful concentrate on their technical habits, deployment necessities, and sensible inference workflows. We start by organising the precise dependencies wanted for Transformers-based execution, verifying GPU availability, and loading openai/gpt-oss-20b with the right configuration utilizing native MXFP4 quantization, torch.bfloat16 activations. As we transfer by means of the tutorial, we work instantly with core capabilities akin to structured era, streaming, multi-turn dialogue dealing with, device execution patterns, and batch inference, whereas preserving in thoughts how open-weight fashions differ from closed-hosted APIs by way…

There’s an previous noticed in administration: What you measure issues. And, sometimes, you get extra of no matter you’re measuring. Software program engineers have debated productiveness metrics for many years, beginning with strains of code. However as the brand new technology of AI coding brokers delivers extra code than ever, what their managers should be measuring is much less clear. Huge token budgets — primarily, the quantity of AI processing energy a developer is permitted to devour — have grow to be a badge of honor amongst Silicon Valley builders, however that’s a really bizarre means to consider productiveness. Measuring…

Nicholas Moore, who pleaded responsible to hacking the U.S. Supreme Court docket’s digital doc submitting system dozens of instances over a number of months, was sentenced on Friday to a yr of probation. Moore had additionally hacked into the community of AmeriCorps, a authorities company that runs stipend volunteer applications, and the methods of the Division of Veterans Affairs, which offers healthcare and welfare to army veterans.  The person bragged about his feats on an Instagram account named @ihackedthegovernment, the place he posted the private info of the individuals he had hacked. Moore used one in all his sufferer’s credentials…

OpenAI is shedding two of the architects of its most bold moonshots. Kevin Weil, who led the corporate’s science analysis initiative, and Invoice Peebles, the researcher behind AI video software Sora, each introduced their departures on Friday. The exits come as OpenAI consolidates round enterprise AI and its forthcoming “superapp.” The departures observe OpenAI’s determination to chop again on “side quests,” together with customer-facing bets like Sora and OpenAI for Science. Sora, which was shedding an estimated $1 million per day in compute prices, was shut down final month. OpenAI for Science was the interior analysis group behind Prism, an…

client = huey.create_consumer( staff=4, worker_type=WORKER_THREAD, periodic=True, initial_delay=0.1, backoff=1.15, max_delay=2.0, scheduler_interval=1, check_worker_health=True, health_check_interval=10, flush_locks=False, ) consumer_thread = threading.Thread(goal=client.run, daemon=True) consumer_thread.begin() print(“Shopper began (threaded).”) print(“nEnqueue fundamentals…”) r1 = quick_add(10, 32) r2 = slow_io(0.75) print(“quick_add outcome:”, r1(blocking=True, timeout=5)) print(“slow_io outcome:”, r2(blocking=True, timeout=5)) print(“nRetries + precedence demo (flaky job)…”) rf = flaky_network_call(p_fail=0.7) strive: print(“flaky_network_call outcome:”, rf(blocking=True, timeout=10)) besides Exception as e: print(“flaky_network_call failed even after retries:”, repr(e)) print(“nContext job (job id inside payload)…”) rp = cpu_pi_estimate(samples=150_000) print(“pi payload:”, rp(blocking=True, timeout=20)) print(“nLocks demo: enqueue a number of locked jobs shortly (ought to serialize)…”) locked_results = [locked_sync_job(tag=f”run{i}”) for i in range(3)] print([res(blocking=True, timeout=10) for res in…

At a classy venue close to the San Francisco pier, Sam Altman’s verification challenge World celebrated its subsequent evolution and fast enlargement of its ambitions.  And it’s beginning with Tinder. Tools for Humanity, the corporate behind the World challenge, introduced Friday plans to combine its verification tech into courting apps, occasion and live performance ticketing methods, enterprise organizations, e-mail, and different arenas of public life. “The world is getting near very highly effective AI, and that is doing numerous fantastic issues,” mentioned Altman, talking earlier than a packed crowd at The Halfway. “We’re additionally heading to a world now the…

What Is AI Crimson Teaming? AI Crimson Teaming is the method of systematically testing synthetic intelligence methods—particularly generative AI and machine studying fashions—towards adversarial assaults and safety stress situations. Crimson teaming goes past basic penetration testing; whereas penetration testing targets recognized software program flaws, crimson teaming probes for unknown AI-specific vulnerabilities, unexpected dangers, and emergent behaviors. The method adopts the mindset of a malicious adversary, simulating assaults comparable to immediate injection, knowledge poisoning, jailbreaking, mannequin evasion, bias exploitation, and knowledge leakage. This ensures AI fashions should not solely sturdy towards conventional threats, but additionally resilient to novel misuse situations distinctive…