Web Design and Development Projects

AI Malware: 97% of Samples Never Reached Production

A new analysis of 405 AI-related malware samples found a much smaller real-world footprint than the volume of public samples suggests. Unit 42, Palo Alto Networks' threat research team, found only 12 of the samples on protected production endpoints, while roughly 97% appeared only in research repositories, sandboxes, or security-testing environments.

That does not make AI malware harmless. It points to a more useful distinction: artificial intelligence is already helping attackers create and package malicious software, but the available evidence does not yet show that AI has made those programs unusually difficult for established security controls to detect.

The 405 samples tell a different story from the headlines

Unit 42 built its dataset from 405 unique SHA-256 hashes gathered from WildFire analysis reports, VirusTotal Intelligence, and published open-source research. The researchers deliberately used a broad definition of AI-related malware, including samples where artificial intelligence was part of the malware itself, its delivery mechanism, or simply its branding. That means the dataset includes genuine malicious code alongside proof-of-concept research and conventional malware disguised as AI software.

When the researchers checked those samples against Palo Alto Networks telemetry, only 12 appeared on non-test Cortex XDR-protected endpoints. That represents 3% of the dataset. Another roughly 15 to 20 unique samples appeared in WildFire network sessions, but the researchers found no evidence that the remaining samples reached customer endpoints or crossed customer firewalls.

There is an important limitation here: this is not a measurement of every AI malware sample circulating on the internet. The endpoint and network telemetry covered activity from December 2024 through June 2025 and came from Palo Alto Networks' own visibility. The 97% figure therefore describes this selected dataset and this telemetry, not the percentage of all AI malware worldwide that is harmless or inactive.

https://wiztechnoz.com/blog/ai-malware-97-of-samples-never-reached-production

Moonshot Wants U.S. Clouds to Host Kimi K3

Moonshot AI is negotiating with Microsoft, Amazon and Google over deals that could put its Kimi K3 model on the biggest U.S. cloud platforms. Reuters reports that the Chinese startup is seeking as much as 30% of revenue generated from K3-related services, although the talks are still early and may not produce agreements. The unusual part is not simply that a Chinese AI company wants access to American cloud infrastructure. It is that Kimi K3's open-weight model may need those clouds precisely because its enormous size makes running it independently difficult for most customers.

Moonshot is negotiating with all three major U.S. clouds

The discussions involve Microsoft's Azure, Amazon Web Services and Google Cloud, according to three people familiar with the talks cited by Reuters. Moonshot is reportedly seeking up to a 30% share of revenue generated through K3 services on those platforms, broadly matching terms it has offered major customers using the model. Microsoft, Google and AWS declined to comment, while Moonshot did not respond to Reuters' request for comment. Because the negotiations are private and unresolved, the proposed revenue split should be treated as a reported negotiating position rather than a finalized commercial arrangement.

If completed, the agreements would create a particularly interesting business model. The cloud provider would supply the computing infrastructure and distribution, while Moonshot would receive a share of the resulting usage revenue. That would give developers access to Kimi K3 without having to purchase and operate the hardware needed to run a model of its scale. It would also give Moonshot a route into enterprise customers that may prefer buying AI through an existing cloud platform rather than building a separate infrastructure stack.

https://wiztechnoz.com/blog/moonshot-wants-us-clouds-to-host-kimi-k3

OpenAI’s Jalapeño Chip Could Make AI Inference Faster

OpenAI has published the first measured results from Jalapeño, its custom chip for running trained AI models, and the numbers point to a different way of improving AI performance: make the hardware itself fit the workload. In tests using public models, OpenAI says Jalapeño delivered between 1.5 and 1.9 times more AI work per watt and between 1.7 and 3.6 times lower end-to-end latency than the comparison systems. The more important story is not simply that OpenAI has designed a chip, but that it is trying to control more of the stack that determines how quickly and efficiently AI reaches users.

Jalapeño targets the part of AI people actually use

Jalapeño is an inference chip, meaning it is designed to run an already-trained model and generate responses rather than train a new model from scratch. That distinction matters because inference happens every time someone asks an AI system a question, runs an agent, or sends a request through an application programming interface. OpenAI says Jalapeño is intended to improve both throughput, or how much useful AI work the system can handle, and latency, or how long users wait for a response. OpenAI first unveiled the chip in June as part of a multigeneration hardware platform being developed with Broadcom and other partners.

https://wiztechnoz.com/blog/openais-jalapeo-chip-could-make-ai-inference-faster

Google’s Gemini Enterprise Takes AI Deeper Into Finance

Financial analysts are getting a new kind of AI assistant: one designed to work with licensed market data, regulatory filings, internal databases, and the controls that banks already depend on. Google Cloud announced Gemini Enterprise for Financial Services on August 25, putting the service into preview for capital markets and corporate banking and positioning it as a more specialized alternative to general-purpose AI tools.

The interesting part is not simply that Google has packaged Gemini for another industry. The platform is built around a Financial Research agent, more than 50 financial skills, 13 data connectors, and governance features intended to make AI output traceable. That combination points to a broader shift in enterprise AI: the valuable system may increasingly be the one that connects a capable model to trusted business data and controlled workflows, rather than the model with the most impressive general benchmark score.

Why general-purpose AI struggles in financial work

A financial analyst rarely works from one clean source. Preparing research can involve market feeds, company information, regulatory documents, internal models, licensed research, and confidential client material. A general-purpose AI model can summarize or reason about these materials, but it does not automatically know which source is authoritative, which data a particular employee may access, or how an institution wants the work documented. Google Cloud argues that those missing pieces are a major barrier to deploying AI safely in regulated financial organizations.

That distinction changes what an enterprise AI product has to solve. Accuracy is only one requirement when an analyst may need to defend an answer to a colleague, client, auditor, or regulator. The system also needs access controls, traceability, current information, and a clear record of where an answer came from. Gemini Enterprise for Financial Services is designed around those requirements rather than treating them as features that customers must build around a generic chatbot.

https://wiztechnoz.com/blog/googles-gemini-enterprise-takes-ai-deeper-into-finance

Why Businesses Are Choosing Cheaper AI Models Over the Best Ones

The Enterprise AI Race Is Starting to Look Different

The biggest AI model is not necessarily the model businesses want to buy. New spending data from Ramp suggests that companies are increasingly choosing AI models based on price, practical performance, and value rather than automatically moving to the newest flagship model.

The clearest example is Anthropic's Fable 5. Despite being positioned as the company's most capable model, Fable 5 accounted for only about 6% of tokens purchased from Anthropic and 11.4% of spending on Anthropic models in its first full month tracked by Ramp.

That result is important because it challenges a basic assumption of the AI market: that customers will naturally pay more for the strongest available model.

https://wiztechnoz.com/blog/why-businesses-are-choosing-cheaper-ai-models-over-the-best-ones

OpenAI Gives GPT-5.6 a New Role in AI Coding Agent Kiro

GPT-5.6 Moves Deeper Into Professional Software Development

OpenAI has expanded the reach of its GPT-5.6 model family by bringing it to Kiro, an AI-powered software development agent designed for teams working through the full engineering lifecycle. The integration was announced on August 24, 2026, putting OpenAI's latest flagship models directly into a workflow built around planning, implementation, review, and testing.

The announcement is significant because it is less about adding another chatbot interface and more about embedding advanced AI models inside a structured development environment. Kiro is designed to turn high-level requirements into technical designs and executable tasks, giving the model more context about what developers are actually trying to build.

https://wiztechnoz.com/blog/openai-gives-gpt-56-a-new-role-in-ai-coding-agent-kiro

Thomson Reuters Builds Its Own AI Model for $40 Million

Thomson Reuters Is Taking a Different Route to Frontier AI

Thomson Reuters has launched Thomson, its first proprietary large language model, in a move that highlights a changing strategy in enterprise artificial intelligence. Instead of trying to compete with the biggest general-purpose models through enormous training budgets, the company started with an open-source foundation and focused its investment on specialized professional knowledge, training, and domain expertise.

The company says it invested about $40 million in talent and computing to develop Thomson. That is a very different scale from the multibillion-dollar infrastructure investments associated with the largest frontier AI labs. Thomson Reuters argues that the approach can produce a highly capable model while giving the company direct control over the technology and its operating costs.

https://wiztechnoz.com/blog/thomson-reuters-builds-its-own-ai-model-for-40-million

Apollo Global Data Breach Exposes Sensitive Personal Data

Apollo Global Confirms Cloud Data Breach After Social Engineering Attack

Apollo Global Management has confirmed a cybersecurity incident in which attackers gained unauthorized access to certain cloud platforms and potentially exposed sensitive personal information. The disclosure, made public on August 21, 2026, places Apollo among a growing group of financial organizations targeted by social-engineering attacks designed to obtain access through employees rather than by exploiting a newly disclosed software vulnerability.

According to Apollo's disclosure, unauthorized access occurred between July 6 and July 10. The company later determined that information potentially affected by the incident included names, dates of birth, contact information, home addresses and Social Security numbers.

https://wiztechnoz.com/blog/apollo-global-data-breach-exposes-sensitive-personal-data

Go 1.27 Is Here: What Developers Need to Know Before Upgrading

Go 1.27 Is More Than a Routine Go Release

Go 1.27 arrived on August 19, bringing one of the more substantial updates to the programming language in recent releases. The release adds generic methods, a redesigned JSON implementation, built-in UUID support, post-quantum cryptography, better goroutine leak detection, faster small-object allocation and experimental SIMD support. :contentReference[oaicite:0]{index=0}

For developers maintaining Go APIs, backend services, CLI tools or infrastructure software, the interesting question is not simply whether to install Go 1.27. It is which changes are worth adopting immediately and which ones deserve testing before they reach production.

https://wiztechnoz.com/blog/go-127-is-here-what-developers-need-to-know-before-upgrading

😀😀😀