AI Won’t Replace Software Engineers—But It Will Change the Job
A coding agent can inspect a repository, edit several files, run tests, and open a pull request. That is a meaningful expansion beyond autocomplete.
Continued
GradCuit: How to Make LLMs Reason Better at Test Time Without Changing a Single Weight
TL;DR GradCuit (arXiv:2608.02585) inserts optimizable latent vectors at an intermediate Transformer layer and uses causal self-attention as a differentiable "circuit" to flow reward-weighted gradients…
Goldman Sachs' Keung on Chinese AI Models
Ronald Keung, Head of Asia Internet Research at Goldman Sachs, expects competition for best performance-to-cost balance to intensify among Chinese AI developers.
I built a website without leaving ChatGPT — and it only took one prompt
I used ChatGPT Sites to turn four book reviews into a polished monthly reading website without writing code. Here’s how it worked and the prompt to try.
AI used new levels of ‘autonomy and deception’ to trick people in safety test
The latest artificial intelligence (AI) tools from Anthropic and OpenAI went to new extremes in trying to undermine a popular platform during testing by the UK's AI Security Institute.
Has the AI race shifted from U.S. vs China to open vs closed?
Whether it’s GLM-5.2, Kimi K3, or DeepSeek V4, China’s open-source AI models are reshaping how we think about AI.
‘I feel like I dug my own grave’: The workers caught in the AI transition
Lisa says she never felt the need to use artificial intelligence (AI) in her work.
Amazon loses US court ban on Perplexity's AI shopping tools
In a novel legal dispute, the San Francisco-based 9th U.S. Circuit Court of Appeals determined that Amazon was unlikely to succeed on its claim that Perplexity's AI agents violated a federal…
Why IPO hype always comes before the fall
Following the hotly hyped debut of SpaceX in America and the booming Shanghai listing of Chinese semiconductor company ChangXin Memory Technologies (CXMT), initial public offerings are being eagerly…
Tencent ramps up overseas AI push with wider rollout of flagship Hy3 model
Tencent Holdings is ramping up its overseas push for its flagship artificial intelligence (AI) model Hy3, joining a wave of Chinese technology rivals and rising labs taking aim at the world’s leading…
China’s AI Agent Unicorn Lovart Is Said to Explore Hong Kong IPO
Evoken, the Chinese startup behind AI design agent Lovart, is considering an initial public offering in Hong Kong, people familiar with the matter said, the latest such firm seeking to tap booming…
Trump advisers tell AI firms they will not safety-test open-weight models
Open models, including Nvidia's Nemotron and Meta's Llama, are AI systems with publicly accessible core components. Closed models are controlled by specific companies. Major U.S.
I Audited My AI's To-Do List. A Quarter of It Was Already Done.
My coding agent has a to-do list. It lives in a public GitHub repo — one issue per task, labelled by project, opened and closed automatically as the agent works.
- AI agent reported 25% of tasks as "open" despite completion
- Manual bookkeeping required to close tasks after completion
- Audit verified task completion via external sources like email
TCS and Infosys think they can profit from AI paranoia
The post TCS and Infosys think they can profit from AI paranoia appeared first on The Ken .
- 60% of Digitate clients prefer locally hosted AI tools
- Data security concerns deter clients from adopting Ignio
- IT-services firms exploit trust gap to offer AI solutions
China’s open-weight AI models to be spared US tests, sources say
Debate over how to handle open-weight AI models has intensified following a wave of such releases from Chinese developers.
ASEAN doesn’t need to win the AI race, it needs to run it together
The world’s AI conversation is dominated by a handful of names: the US hyperscalers, the Chinese national champions, and a few European regulatory bodies trying to keep up.
Manipal Health Set for Flat Mumbai Debut After $960 Million IPO
Manipal Health Enterprises Ltd. is set for a muted stock market debut in Mumbai after its $960 million initial public offering, as gray market trading signals cooling investor sentiment.
SpaceX shares sink after execs promise more spending
Still, shares wobbled after an initial jump on Tuesday.
AI researchers let models off the leash – then watched as they tried to add malware to a FOSS project
Models used social engineering and collaborated among themselves to solve a security challenge
Nvidia bets on Sarvam AI; Sidbi eases fund rules for VCs
Happy Wednesday! Sarvam’s Pratyush Kumar has confirmed adding Nvidia to the AI startup’s captable. This and more in today’s ETtech Morning Dispatch.
The new border: Why server farms are the battleground of AI sovereignty
In May, US authorities named a Bangkok firm tied to Thailand’s national AI push, accusing it of helping move billions in Nvidia-powered servers to Chinese buyers, including Alibaba.
SpaceX's AI spending unnerves Wall Street despite promises of quick payoff
In addressing Wall Street for the first time since SpaceX's IPO, executives tried to make the case that hefty AI investments are paying off within a year.
White House, AI firms keep safety framework talks private
The White House met with representatives from leading artificial intelligence companies today to discuss a safety framework for the government to review frontier models prior to launch, although…
SoftBank jumps 10% as Asia tech stocks track Wall Street AI rally
Asian technology stocks rallied on Wednesday after Wall Street's record-setting session fueled optimism around artificial intelligence and growth stocks.
Turning AI momentum into business transformation
As organisations move from AI experimentation to enterprise-wide adoption, SAP’s latest research highlights what businesses need to maximise AI’s full value.
Washington banned Mythos and Fable: It created a hydra
The lesson from Washington’s intervention against Anthropic’s Fable 5 and Mythos 5 is not that governments are powerless over AI. They are not. A state can order a company to switch off a model.
- U.S. government halts release of Anthropic's Fable 5 and Mythos 5 models
- Model ban incentivizes alternative AI development, creating a hydra effect
- Falling costs and rapid catch-up phenomenon make AI capabilities hard to contain
I tried Gemini Spark in Chrome and it's the perfect AI for handling all the boring bits online
Gemini Spark is designed to take on longer-running tasks that continue in the background and now it is part of the Chrome browser, so here's how it works.
Arista Networks’ stock jumps on stellar earnings and revenue beat and strong forecast
Shares of Arista Networks Inc. soared in extended trading today after the networking infrastructure firm easily beat Wall Street’s expectations in its second-quarter earnings report.
- Arista Networks' stock jumps 11% on strong earnings and revenue beat
- Company reports $1.02 earnings per share, surpassing analyst forecast
- CEO credits 2.0 platform strategy for success and strong guidance
KI: Another incident – KI sent phishing emails
According to British researchers, an Anthropic AI tried to manipulate a human via email during a test. The incident fuels further concerns about AI cyberattacks.
Artificial Intelligence: Next Alarm: AI Sent People Phishing Emails
To solve a test task, artificial intelligence attempted to infect publicly accessible software without planning - and to manipulate people with phishing emails. Researchers are sounding the alarm.
AMD data centre sales double but AI chip giant’s shares sink on lofty expectations
Its stock had more than doubled this year after the company emerged as a key contender to Nvidia in AI processors.
- AMD's data center sales double to US$6.7 billion
- Stock price drops 8.8% post-market trading
- Investors set higher expectations for AI chip performance
AI used new levels of 'autonomy and deception' to trick people in safety test
The UK's AI Safety Institute said recent behaviour from Anthropic and OpenAI models was malicious and unprecedented.
Hoike: A Joint-Embedding Predictive Architecture for Transcriptome Data Generation with Diffusion Models
In biomarker discovery, access to sufficient quantities of condition-specific transcriptomic data is often limited by cohort size, privacy concerns, and domain shift between normal and condition…
G-quadruplex structures act as a novel recognition motif for the meiosis-specific histone methyltransferase PRDM9
The histone methyltransferase PR domain containing protein 9 (PRDM9) is a key determinant of meiotic recombination in humans.
MEGA-ODE: Learning Biologically Structured and Navigable Continuous Perturbation Dynamics from Sparse Omics
Perturbation-omics experiments usually measure only a subset of molecular feature, intervention and time space, leaving many response trajectories, perturbation effects and disease- or…
AI Influencers Are Heading Into Uncharted Territory
Some creators fear the EU AI Act’s regulatory chaos will upend their lucrative businesses. Others are owning it by incorporating AI transparency into their creative process.
Trajectory forecasting lifts the femtosecond ceiling in molecular simulation
Nature Machine Intelligence, Published online: 05 August 2026; doi:10.1038/s42256-026-01275-z A new deep learning method enables molecular dynamics simulations over longer time scales while still…
Sliding mode control of robot manipulators via an improved recurrent neural network and barrier functions
Scientific Reports, Published online: 05 August 2026; doi:10.1038/s41598-026-65037-y Sliding mode control of robot manipulators via an improved recurrent neural network and barrier functions
Development and external validation of a nomogram for overall survival in oral tongue squamous cell carcinoma using machine learning-assisted feature selection
Scientific Reports, Published online: 05 August 2026; doi:10.1038/s41598-026-65230-z Development and external validation of a nomogram for overall survival in oral tongue squamous cell carcinoma using…
Clinical-guided deep learning framework for diabetic retinopathy: integrating lesion-aware attention, adversarial augmentation, and uncertainty quantification
Scientific Reports, Published online: 05 August 2026; doi:10.1038/s41598-026-58690-w Clinical-guided deep learning framework for diabetic retinopathy: integrating lesion-aware attention, adversarial…
Apo kobar sumo?
Tidak mustahil jika sudah ada orang Negeri Sembilan mula berkata 'kamu ajak kita bersatu kamu sendiri tak bersatu'.
New release of LLM adds support for reasoning traces, OpenAI Responses, server-side tools, and smarter logging
I released LLM 0.32 this morning, the most significant new version of LLM since the initial launch of the project. The new version includes support for visible reasoning traces, server-side provider…
- Version 0.32 of LLM released on August 4, 2026
- Adds reasoning traces, server-side tools, content-addressable logs
- llm-anthropic plugin enhanced with WebSearch, WebFetch, CodeExecution
AMD’s AI engine shifts into higher gear as data center revenue more than doubles, Helios ramps & market is confused
Advanced Micro Devices Inc. delivered another strong quarter Tuesday, beating Wall Street expectations as its artificial intelligence infrastructure business continued its rapid expansion.
These are the AI world's biggest questions as the White House hashes out a framework with OpenAI, Google, and Anthropic
OpenAI, Anthropic, and Google met in D.C. to review Trump's AI oversight framework amid cybersecurity concerns and regulatory questions.
Outback town says grid connection key to capturing AI investment
An outback town says it has the land and space for AI infrastructure, but without a connection to the national electricity grid, investment remains on hold.
W’Bank urges developing nations to accelerate AI adoption
The World Bank urges developing nations to accelerate AI adoption to boost economic growth and public services, highlighting its potential to bridge develo Read More…
SoftBank Earnings to Test Appetite for AI Bets Beyond ChatGPT
SoftBank Group Corp.’s stock faces a critical test this week as investors seek reassurance that the company’s AI value extends beyond its debt-fueled bet on embattled ChatGPT operator OpenAI.
OK, Well, Rogue AI Agents Are Hacking Again
Rogue AI agents from OpenAI and Anthropic have again been caught trying to disrupt servers and software—and leaving instructions for future bad behavior.
The Anti-AI Backlash Against Hank Green, Explained
YouTuber Hank Green sparked fan backlash after admitting to using AI to research videos—here's the controversy, explained.
An LLM agent attempts to compromise a project on GitHub
The AI Security Institute has released a detailed report on an security incident of its own making. The Institute set some LLM agents loose on the Internet with a security challenge; soon they were…
- LLM agents created security challenge for testing
- Agent opened malicious pull request in REPOA
- Agent sent five emails with malware attempts
AI-generated stories rated better quality than human-written ones, study finds
Researcher says simpler writing by AI is easier to read and digest, but this does not mean human authors are obsolete From Homer’s Odyssey to Agatha Christie’s crime fiction, humans have produced…
Most people prefer AI-written stories – and can’t tell when they are made by humans
People appear to prefer short stories created by AI over those written by humans. They are also no better than chance at telling which is which
‘Baffling’: White House won’t publicly release AI model evaluation framework it reviewed today with OpenAI, Anthropic, Microsoft and others
It's unclear why the administration is keeping the framework under wraps, especially following a series of hacks from OpenAI and Anthropic that have spooked the public.
'I feel like I dug my own grave': The workers caught in the AI transition
AI is reshaping the Philippines' outsourcing industry, raising questions over the future of the industry.
AI fuels more than half of cybercrime in Africa as scams surge – Interpol
https://www.interpol.int/Media/Documents/Publications/Cyberc... Comments URL: https://news.ycombinator.com/item?id=49175826 Points: 268 # Comments: 211
llm-anthropic 0.26
Release: llm-anthropic 0.26 Includes new features enabled by LLM 0.32 : New models: claude-fable-5 , claude-sonnet-5 , and claude-opus-5 .
AI creates psychiatry training cases, but human oversight remains essential
As the use of artificial intelligence for mental health conditions grows among both patients seeking support and psychiatrists treating them, West Virginia University researchers put ChatGPT-5 Pro to the test to determine its abilities in psychiatry education.
SpaceX's first public earnings statement shows the financials of an AI company in 2026
That's a lot of capital expenditure.