Quote for the day:
"Success is not the absence of failure; it's the persistence through failure." -- Aisha Tyle
How cybersecurity leaders can defend against the spur of AI-driven NHI
Many companies don’t have lifecycle management for all their machine identities
and security teams may be reluctant to shut down old accounts because doing so
might break critical business processes. ... Access-management systems that
provide one-time-use credentials to be used exactly when they are needed are
cumbersome to set up. And some systems come with default logins like “admin”
that are never changed. ... AI agents are the next step in the evolution of
generative AI. Unlike chatbots, which only work with company data when provided
by a user or an augmented prompt, agents are typically more autonomous, and can
go out and find needed information on their own. This means that they need
access to enterprise systems, at a level that would allow them to carry out all
their assigned tasks. “The thing I’m worried about first is misconfiguration,”
says Yageo’s Taylor. If an AI agent’s permissions are set incorrectly “it opens
up the door to a lot of bad things to happen.” Because of their ability to plan,
reason, act, and learn AI agents can exhibit unpredictable and emergent
behaviors. An AI agent that’s been instructed to accomplish a particular goal
might find a way to do it in an unanticipated way, and with unanticipated
consequences. This risk is magnified even further, with agentic AI systems that
use multiple AI agents working together to complete bigger tasks, or even
automate entire business processes. The silent backbone of 5G & beyond: How network APIs are powering the future of connectivity
Network APIs are fueling a transformation by making telecom networks
programmable and monetisable platforms that accelerate innovation, improve
customer experiences, and open new revenue streams. ... Contextual
intelligence is what makes these new-generation APIs so attractive. Your needs
change significantly depending on whether you’re playing a cloud game,
streaming a match, or participating in a remote meeting. Programmable networks
can now detect these needs and adjust dynamically. Take the example of a user
streaming a football match. With network APIs, a telecom operator can offer
temporary bandwidth boosts just for the game’s duration. Once it ends, the
network automatically reverts to the user’s standard plan—no friction, no
intervention. ... Programmable networks are expected to have the greatest
impact in Industry 4.0, which goes beyond consumer applications. ... 5G
combined IOT and with network APIs enables industrial systems to become truly
connected and intelligent. Remote monitoring of manufacturing equipment allows
for real-time maintenance schedule adjustments based on machine behavior. Over
a programmable, secure network, an API-triggered alert can coordinate a remote
diagnostic session and even start remedial actions if a fault is found.Quantum Computers Just Reached the Holy Grail – No Assumptions, No Limits
A breakthrough led by Daniel Lidar, a professor of engineering at USC and an
expert in quantum error correction, has pushed quantum computing past a key
milestone. Working with researchers from USC and Johns Hopkins, Lidar’s team
demonstrated a powerful exponential speedup using two of IBM’s 127-qubit Eagle
quantum processors — all operated remotely through the cloud. Their results were
published in the prestigious journal Physical Review X. “There have previously
been demonstrations of more modest types of speedups like a polynomial speedup,
says Lidar, who is also the cofounder of Quantum Elements, Inc. “But an
exponential speedup is the most dramatic type of speed up that we expect to see
from quantum computers.” ... What makes a speedup “unconditional,” Lidar
explains, is that it doesn’t rely on any unproven assumptions. Prior speedup
claims required the assumption that there is no better classical algorithm
against which to benchmark the quantum algorithm. Here, the team led by Lidar
used an algorithm they modified for the quantum computer to solve a variation of
“Simon’s problem,” an early example of quantum algorithms that can, in theory,
solve a task exponentially faster than any classical counterpart,
unconditionally.
4 things that make an AI strategy work in the short and long term
Most AI gains came from embedding tools like Microsoft Copilot, GitHub Copilot,
and OpenAI APIs into existing workflows. Aviad Almagor, VP of technology
innovation at tech company Trimble, also notes that more than 90% of Trimble
engineers use Github Copilot. The ROI, he says, is evident in shorter
development cycles, and reduced friction in HR and customer service. Moreover,
Trimble has introduced AI into their transportation management system, where AI
agents optimize freight procurement by dynamically matching shippers and
carriers. ... While analysts often lament the difficulty of showing short-term
ROI for AI projects, these four organizations disagree — at least in part. Their
secret: flexible thinking and diverse metrics. They view ROI not only as dollars
saved or earned, but also as time saved, satisfaction increased, and strategic
flexibility gained. London says that Upwave listens for customer signals like
positive feedback, contract renewals, and increased engagement with AI-generated
content. Given the low cost of implementing prebuilt AI models, even modest wins
yield high returns. For example, if a customer cites an AI-generated feature as
a reason to renew or expand their contract, that’s taken as a strong ROI
indicator. Trimble uses lifecycle metrics in engineering and operations. For
instance, one customer used Trimble AI tools to reduce the time it took to
perform a tunnel safety analysis from 30 minutes to just three.
How IT Leaders Can Rise to a CIO or Other C-level Position
For any IT professional who aspires to become a CIO, the key is to start
thinking like a business leader, not just a technologist, says Antony Marceles,
a technology consultant and founder of software staffing firm Pumex. "This means
taking every opportunity to understand the why behind the technology, how it
impacts revenue, operations, and customer experience," he explained in an email.
The most successful tech leaders aren't necessarily great technical experts, but
they possess the ability to translate tech speak into business strategy,
Marceles says, adding that "Volunteering for cross-functional projects and
asking to sit in on executive discussions can give you that perspective." ...
CIOs rarely have solo success stories; they're built up by the teams around
them, Marceles says. "Colleagues can support a future CIO by giving honest
feedback, nominating them for opportunities, and looping them into strategic
conversations." Networking also plays a pivotal role in career advancement, not
just for exposure, but for learning how other organizations approach IT
leadership, he adds. Don't underestimate the power of having an executive
sponsor, someone who can speak to your capabilities when you’re not there to
speak for yourself, Eidem says. "The combination of delivering value and having
someone champion that value -- that's what creates real upward momentum."
SLMs vs. LLMs: Efficiency and adaptability take centre stage
SLMs are becoming central to Agentic AI systems due to their inherent efficiency and adaptability. Agentic AI systems typically involve multiple autonomous agents that collaborate on complex, multi-step tasks and interact with environments. Fine-tuning methods like Reinforcement Learning (RL) effectively imbue SLMs with task-specific knowledge and external tool-use capabilities, which are crucial for agentic operations. This enables SLMs to be efficiently deployed for real-time interactions and adaptive workflow automation, overcoming the prohibitive costs and latency often associated with larger models in agentic contexts. ... Operating entirely on-premises ensures that decisions are made instantly at the data source, eliminating network delays and safeguarding sensitive information. This enables timely interpretation of equipment alerts, detection of inventory issues, and real-time workflow adjustments, supporting faster and more secure enterprise operations. SLMs also enable real-time reasoning and decision-making through advanced fine-tuning, especially Reinforcement Learning. RL allows SLMs to learn from verifiable rewards, teaching them to reason through complex problems, choose optimal paths, and effectively use external tools.Quantum’s quandary: racing toward reality or stuck in hyperbole?
One important reason is for researchers to demonstrate their advances and show
that they are adding value. Quantum computing research requires significant
expenditure, and the return on investment will be substantial if a quantum
computer can solve problems previously deemed unsolvable. However, this return
is not assured, nor is the timeframe for when a useful quantum computer might be
achievable. To continue to receive funding and backing for what ultimately is a
gamble, researchers need to show progress — to their bosses, investors, and
stakeholders. ... As soon as such announcements are made, scientists and
researchers scrutinize them for weaknesses and hyperbole. The benchmarks used
for these tests are subject to immense debate, with many critics arguing that
the computations are not practical problems or that success in one problem does
not imply broader applicability. In Microsoft’s case, a lack of peer-reviewed
data means there is uncertainty about whether the Majorana particle even exists
beyond theory. The scientific method encourages debate and repetition, with the
aim of reaching a consensus on what is true. However, in quantum computing,
marketing hype and the need to demonstrate advancement take priority over the
verification of claims, making it difficult to place these announcements in the
context of the bigger picture.
Ethical AI for Product Owners and Product Managers
Sharded vs. Distributed: The Math Behind Resilience and High Availability
In probability theory, independent events are events whose outcomes do not
affect each other. For example, when throwing four dice, the number displayed on
each dice is independent of the other three dice. Similarly, the availability of
each server in a six-node application-sharded cluster is independent of the
others. This means that each server has an individual probability of being
available or unavailable, and the failure of one server is not affected by the
failure or otherwise of other servers in the cluster. In reality, there may be
shared resources or shared infrastructure that links the availability of one
server to another. In mathematical terms, this means that the events are
dependent. However, we consider the probability of these types of failures to be
low, and therefore, we do not take them into account in this analysis. ...
Traditional architectures are limited by single-node failure risk.
Application-level sharding compounds this problem because if any node goes down,
its shard and therefore the total system becomes unavailable. In contrast,
distributed databases with quorum-based consensus (like YugabyteDB) provide
fault tolerance and scalability, enabling higher resilience and improved
availability.
