Ripple Energy has announced that it is creating the UK’s first c-owned solar park, allowing those who buy shares to benefit from green electricity, money off bills, and the feelgood factor of the climate and environmental benefits.

Where?

Ripple Energy says the new ‘Derril Water Solar Park’ will be on a 175-acre site in the Torridge district of Devon, 1.2km southwest of the village of Pyworthy.

What? 

The new solar park will have about 70,000 solar panels, giving it the potential to power around 14,000 homes with cheap green power.

When? 

Ripple says that the co-ownership share offer, which is open now, will close in mid-May and the park will be operational by late summer 2024.

How? 

People can buy shares in the cooperative society that will own part of the Derril Water Solar Park with shares starting at £25. The co-owner’s electricity supplier then buys their share of the solar park’s electricity and supplies it to their home or premises via the grid and automatically supplies the savings to their electricity bill.

The savings will be equal to the wholesale value of the power generated minus the operating cost of the solar park, multiplied by the co-owner’s share of the generation.

How Much? 

Ripple says, for a typical home, the cost is around £3,000 (paid on application) which buys enough of the solar park to generate 100 per cent of their electricity needs for 40 years. The minimum ownership is £25, and the maximum owners can invest is what would meet 120 per cent of their electricity needs. Ripple estimates that the effective rate of return will be 5 per cent over the 40-year operational life of the solar park with a member paying about £3,000 to get £170 as savings on their electricity bills a year.

The Benefits 

Some of the many anticipated benefits of the solar park include:

– The environmental and climate benefits of a saving of 19,000 tonnes of CO2 per year which is the equivalent of taking 12,250 cars off the road or the work of 6,785,000 trees.

– Savings on electricity bills and providing a cheaper alternative to solar panels for the house (up to 50 per cent cheaper than rooftop solar)

– The feel-good factor for co-owners of knowing that they’re reducing their carbon footprint as well as directly increasing the UK’s renewable energy capacity.

– It’s a fully portable scheme i.e., co-owners can take it with them if they move.

– Ripple says other benefits will come from the site acting as a host for ecological and biodiversity enhancements e.g., new bee banks, wildflower meadows and hedgerows.

What Does This Mean For Your Organisation? 

This co-ownership project is an example how the demand for green energy in addition to cost savings is creating new options for consumers – this is a co-ownership project rather than a green tariff. As Ripple points out, land use today has become more of a balance between ensuring food security, maintaining biodiversity, infrastructure demands and addressing the climate crisis. This type of scheme, a first in the UK, looks like it could benefit domestic and business co-owners by meeting a number of important needs and could be the shape of things to come as solar parks and the solar industry in the UK are on a growth path to help the UK meet its green targets.

The Birth of The IC

Around this time 62 years ago (April 25th, 1961), Robert Noyce (co-founder of Fairchild Semiconductor) patented his invention – a monolithic integrated circuit (IC).

Jack Kilby patented a similar invention, but his version was difficult to mass produce due to external gold wire connections. Thus, Noyce’s version of the IC became widely adopted due to being more practical.

This was the first step in making computers more affordable as the new ICs reduced production costs significantly, increased efficiency and allowed for production of more powerful machines. To this day, modern chips are based on the ICs developed by Noyce.

The concept of combining multiple components in one electrical device was not new at the time. Interestingly, one of the first attempts to do so was a vacuum tube from the 1920s. It was designed for the purpose of tax avoidance in Germany, as radio receivers were taxed based on how many tube holders the receiver had.

Noyce would go on to co-found Intel. He was nicknamed the ‘Mayor of Silicon Valley’ as one of the first scientists to work there. And, because both Fairchild Semiconductor and Intel had great impact on the silicon industry.

Under Noyce’s leadership, yet another breakthrough took place at Intel. The creation of the microprocessor. He believed in a management style that promoted teamwork and sharing of ideas and was not afraid to take calculated risks. This paid off greatly, when in 1969, Marcian Hoff (a recent hire) approached him with concerns about a new design for Busicom’s electronic calculator.

That year, Intel was approached by Busicom to create a new design for their calculator. Hoff was concerned that the current design would make meeting price goals impossible. Instead, they should combine the chips to reduce complexity and costs. Which resulted in the Intel 4004, the first commercial microprocessor.

Intel went on to become a key player in the computer industry. Around 80% of personal computers in the world run on an Intel processor, meaning you are probably reading this article on a device with one and currently, Intel is valued at a whopping $128.72 billion.

Robert Noyce has left a gargantuan mark on the IT industry and the world.

If you’ve produced some content using ChatGPT and, knowing that it can sometimes generate incorrect information, would quickly like to check that the source data is from reputable sites, here’s how:

– In the window below the last response in a chat, simply type “please provide sources for the information you presented in this response” (or similar).

– ChatGPT will list the sources/links of the web pages it visited to get the information.

– If you’d like more links on the same subject, input an instruction like “list some popular links related to (subject matter here…).”

Amazon has announced the launch of ‘Amazon Bedrock,’ a new service which it says is the easiest way for customers to build and scale generative AI-based applications using foundation models (FMs).

What Is A Foundation Model (FM)? 

A foundation model is an “ultra-large” machine learning model that generative AI uses. FMs, which are machine learning (ML) models, are pre-trained on vast amounts of data and power generative AI. The term generative AI refers to a type of AI that can create new content and ideas, including conversations, stories, images, videos, and music. ChatGTP, for example, is currently the most well-known generative AI chatbot.

Bedrock 

Bedrock is essentially an AI app-building toolkit that’s available to Amazon Web Services (AWS) customers. AWS is the biggest cloud provider in the world and with Bedrock, customers can choose from a range of FMs such as AI21’s Jurassic-2, Anthropic’s Claude, Stability AI’s Stable Diffusion, and Amazon Titan. Using Bedrock (accessible via an API), AWS customers can customise and fine tune a model to perform a specific task without having to annotate large volumes of data.

Easily Find And Customise The Right Model For Your Own Purposes 

Amazon says that with Bedrock’s serverless experience, customers can “easily find the right model for what they’re trying to get done, get started quickly, privately customise FMs with their own data, and easily integrate and deploy them into their applications using the AWS tools and capabilities they are familiar with (including integrations with Amazon SageMaker ML features like Experiments to test different models and Pipelines to manage their FMs at scale) without having to manage any infrastructure.” 

Example 

One example (from AWS) of how customers could use Bedrock is a content marketing manager working for a fashion retailer, needing to develop fresh, targeted ad and campaign copy for an upcoming new line of handbags. AWS says the marketing manager could provide Bedrock a few labelled examples of their best performing taglines from past campaigns, plus the associated product descriptions, then Bedrock could automatically start generating effective social media, display ad, and web copy for the new handbags. However, Amazon is quick to point out that Bedrock would also not include any of the customer’s data to train the underlying models, and with all data being encrypted and not actually leaving a customer’s Virtual Private Cloud (VPC), AWS says customers can trust that their data will remain private and confidential.

The Advantages of Bedrock 

The advantages of Bedrock are:

– It’s relatively straightforward and easy to use and customers can take the base FM and build differentiated apps using their own data (a little data or a lot) just by using a few prompts.

– Customers can access high-performing FMs that give good results and are best-suited for their purposes.

– Bedrock enables a seamless move into applications, without having to manage huge clusters of infrastructure or incur large costs.

– Bedrock allows customisation but keeps the customer’s data secure, and private.

– It’s a scalable and reliable service that uses AWS managed service (the biggest cloud service).

– Bedrock makes the power of FMs accessible to companies of all sizes, thereby allowing them to accelerate the use of ML and get the value from ML across their organisations and make it easy to build their own generative AI applications. Amazon says it thinks “Bedrock will be a massive step forward in democratising FMs.” 

Also Announced 

As well as announcing the introduction of Bedrock, Amazon has also announced:

– The general availability of Amazon EC2 Trn1n instances powered by AWS Trainium and Amazon EC2 Inf2 instances powered by AWS Inferentia2, which it describes as “the most cost-effective cloud infrastructure for generative AI”.

– The general availability of Amazon CodeWhisperer, free for individual developers. Amazon CodeWhisperer is an AI coding companion that uses a FM to improve developer productivity by generating code suggestions in real-time based on developers’ comments in natural language and prior code in their Integrated Development Environment (IDE).

What Does This Mean For Your Business? 

Amazon says AI and ML have been a focus for over 20 years and that many of the capabilities customers use with Amazon are already driven by ML, however Bedrock provides a way for Amazon to compete in a rapidly growing and competitive generative AI (chatbot) market. With Bedrock, Amazon joins ChatGPT (OpenAI), Bard (Google), Copilot (Microsoft), Tongyi Oiawen (Alibaba), and others who are offering businesses the chance to create more value, save costs and time by leveraging the considerable power of AI. Amazon’s AWS is already the biggest cloud service so AWS can take advantage of this to introduce Bedrock to its many business customers, adding value for them and a helping to retain them. For businesses and developers who are AWS customers, Bedrock provides an easy, low-cost way to access and use powerful FMs to create specific, customised tools that can add value to their businesses and for their customers.

In this insight, we look at the ‘Jailbreaking’ concept in ChatGPT and other LLMs, and at what steps can be taken to mitigate the risks to users.

Jailbreaking 

Jailbreaking, in general, refers to the process of removing restrictions or limitations imposed by a device or software, often to gain access to features or functionality that were previously unavailable. One popular example is removing the artificial limitations in iPhones to enable users to install apps not approved by Apple.

Jailbreaking ChatGPT 

In the context of chatbots, and specifically ChatGPT, jailbreaking refers to designing prompts to make ChatGPT bypass its rules or restrictions that are in place to prevent the chatbot from producing hateful or illegal content.

Some security researchers, technologists, and computer scientists have recently reported developing successful jailbreaks and prompt injection attacks against ChatGPT and other generative AI systems to show how vulnerable these systems are. For example, Adversa AI’s CEO Alex Polyakov has reported taking just a couple of hours to break GPT-4 (the latest version of the GPT – Generative Pre-trained Transformer – series of language models from OpenAI). Polyakov has also reported making a “Universal LLM Jailbreak,” which works on many different large language models (LLMs) including OpenAI’s GPT-4, Microsoft’s Bing chat system, Google’s Bard, and Anthropic’s Claude. The jailbreak can fool systems into generating detailed instructions on creating meth and how to hotwire a car!

Prompt Injection Attacks 

Prompt injection attacks, which are related to jailbreaking, can be used to insert malicious data or instructions into AI models in various ways, depending on the specific type of AI model and the nature of the attack. Examples include:

– Language Models. As in the case of ChatGPT, language models are vulnerable to prompt injection attacks where the attacker inputs carefully crafted prompts to generate outputs that may contain hate speech or biased language. In this case, the attacker can inject prompts that contain malicious content such as hate speech or illegal instructions.

– Image Recognition Models. In image recognition models, the attacker can inject malicious data by modifying or manipulating the images used for training the model. For example, an attacker could add hidden messages or images to the training data to alter the model’s behaviour.

– Recommender Systems. In recommender systems, an attacker could inject malicious data by creating fake user accounts and manipulating the recommendations generated by the system. For example, an attacker could create fake accounts and provide biased ratings and feedback to manipulate the recommendations generated by the system.

– Autonomous Systems. In autonomous systems such as self-driving cars, an attacker could inject malicious instructions or data by exploiting vulnerabilities in the software or hardware used by the system. For example, an attacker could hack into the car’s computer system and inject malicious code to alter its behaviour.

Ethical and Security Issues Highlighted 

In a blog post on the Adversa AI’s website about his research, Polyakov highlighted the fact that being able to deploy a Universal LLM Jailbreak raises the issues of:

– Ethics and AI Safety. Ensuring responsible usage is crucial to preventing malicious applications and safeguard user privacy. Also, enterprises implementing LLM’s who fully trust vendors may not be aware of potential issues.

– Demonstrating such jailbreaks shows a fundamental security vulnerability of LLM’s to logic manipulation, whether it’s jailbreaks, Prompt injection attacks, adversarial examples, or any other existing and new ways to hack AI.

How To Prevent Jailbreaks and Prompt Injection Attacks In ChatGPT 

As an AI language model, ChatGPT is designed to be secure and robust, but there is always a possibility that malicious actors could attempt to jailbreak it. Here are a few strategies that can help prevent and mitigate the effects of jailbreaks in ChatGPT:

– Increase awareness of the threat. Increasing awareness and assessing AI-related threats are important first steps in protecting businesses.

– Implement Robust Security Measures during the development. Developers and users of LLMs must prioritise security to protect against potential threats.

– Secure Development. One of the most effective ways to prevent jailbreaks is to ensure that ChatGPT is developed using secure coding practices. This includes minimising the use of vulnerable libraries and APIs, implementing secure authentication and access controls, and regularly testing the system for security vulnerabilities. Assessment and ‘AI Red Teaming’ of models and applications before their release is an important step in mitigating potential issues around jailbreaking.

– Monitoring and Detection. Regular monitoring and detection of suspicious behaviour can help prevent jailbreaks by enabling the identification of unusual activity and the prompt response to security incidents. This includes implementing security monitoring tools and conducting regular security audits to identify potential vulnerabilities.

– Restricted Access. Restricting access to ChatGPT can help prevent jailbreaks by limiting the number of people who have access to the system and minimising the risk of insider threats. This includes implementing strong access controls, using multi-factor authentication, and restricting access to sensitive data and features.

– Regular Updates. Regular updates and patches can help prevent jailbreaks by addressing known vulnerabilities and updating security measures to address new threats. This includes keeping the system up to date with the latest security patches and regularly reviewing and updating security protocols and procedures.

– It’s also important to ensure that ChatGPT, going forward, is developed with ethical considerations in mind and that it adheres to responsible AI principles. This includes considering the potential impacts of the technology on society and taking steps to promote fairness, accountability, and transparency in its usage.

– AI Hardening. As highlighted by Polyakov’s research findings, organisations developing AI technologies should implement extra measures to “harden” AI models and algorithms. These could include adversarial training and more advanced filtering.

What Does This Mean For Your Organisation? 

Although the development of universal jailbreaks as part of research has value in finding vulnerabilities in LLM models and explaining and understanding how they work, it has also highlighted some serious threats and issues. It is a little scary how quickly and easily researchers were able to develop universal jailbreaks for the most popular AI chatbots and demonstrated how these current vulnerabilities could pose safety, security, and potentially legal threats to users as well as raising ethical concerns for LLM developers and highlighting how much more work they need to do quickly to mitigate the risks caused by misuse of their products.  Although worries about threats to privacy from LLMs have already made the news, raising awareness about jailbreaking of these models is another step towards making them safer to use.

A humanoid robotics company (1X Technologies) has announced that it raised $2.5m Series A2 funding led by ChatGPT’s creators OpenAI to pursue producing androids/humanoid robots at commercial scale.

Funding 

The funding round to develop more androids was led by the OpenAI Startup Fund, with participation from Tiger Global and a consortium of Norway-based investors, including Sandwater, Alliance Ventures, and Skagerak Capital.

1X Technologies 

Norwegian company 1X Technologies (formerly Halodi Robotics) founded in 2014, makes androids capable of human-like movements and behaviours, and its newest android iteration ‘NEO’ is powered by artificial intelligence but takes the form of a human-like body, i.e. an upright bi-pedal android with arms and legs.

The company previously made the news for its ‘EVE’ android – a teleoperated humanoid robot which has a human shape with arms, bending legs and a head, yet moves on a wheeled platform. Like the newest ‘Neo’ robot, EVE can perform many human tasks with its arms, such as packing goods.

No Robot Butler Just Yet 

Those hoping to have an android robot butler in the near future thanks to the new funding round, however, will be disappointed as 1X Technologies mission and focus is to create robots with practical, real-world applications to augment labour globally.

Money To Scale-Up

As well as increasing its efforts in building the upcoming bipedal android model Neo, the new funding will enable 1X Technologies to scale-up manufacturing of its first commercially available android EVE in Norway and North America.

Android Robots For Automated Labour 

IX’s Brad Lightcap says that the company is “at the forefront of augmenting labour through the use of safe, advanced technologies in robotics,” and OpenAI’s COO and manager of the OpenAI Startup Fund also confirmed that the new androids will initially be focused on augmenting the labour force by saying “The OpenAI Startup Fund believes in the approach and impact that 1X can have on the future of work.” 

Necessary To Try Them Out In Real World Settings 

Bernt Øyvind Børnich CEO and founder of 1X Technologies has highlighted the need to get androids deployed in the real world to learn lessons that could improve their effectiveness saying: “Deploying our wheeled android EVE at an unprecedented commercial scale gives us a unique understanding of the challenges and opportunities the robotics community has yet to address. If androids are going to work in our world, they need to experience our world.” 

Plug A Labour Gap 

Arne Tonning, Partner, Alliance Venture (one of the companies involved in the funding round) has highlighted how androids could also serve a useful purpose in plugging a labour gap. For example, Tonning said on the 1X technologies website: “Demographic changes will cause a labour shortage, and androids could help fill the gap. Goldman Sachs predicts a $150 Billion US market potential in 2035. Solving the right use cases is key to success, and we believe 1X Technologies is constructing a winning alliance.” 

It is anticipated that humanoid robots may help fill this gap by augmenting the labour market with their ability to perform repetitive, dangerous tasks or tasks that require high precision.

Others 

1X Technologies are by no means the only company working on creating androids focused on augmenting the labour market. There are several companies working on humanoid robots designed to work in various industries. Examples include:

– Boston Dynamics. Known for their advanced robots, Boston Dynamics has developed humanoid robots like Atlas and Spot that are designed to perform various tasks, including industrial inspections, construction work, and search and rescue operations.

– SoftBank Robotics. SoftBank Robotics is a Japanese company that has developed humanoid robots like Pepper and NAO, which are designed to interact with humans and assist with tasks like customer service, education, and healthcare.

– Hanson Robotics. Hanson Robotics is a Hong Kong-based company that has developed humanoid robots like Sophia, which can understand and respond to human speech and facial expressions and is designed for use in applications like customer service, education, and entertainment.

– PAL Robotics. PAL Robotics is a Spanish company that has developed humanoid robots like REEM-C and TALOS, which are designed to work in industries like logistics, healthcare, and hospitality.

– Toyota Robotics. Toyota Robotics is a subsidiary of the Toyota Motor Corporation and it is working on developing humanoid robots like T-HR3, which can be remotely controlled by humans and is designed for use in applications like disaster response and home assistance.

Amazon and Tesla? 

Although Amazon and Tesla are not primarily known for developing humanoid robots, yet they have both invested in developing robotic technology for various purposes, particularly Amazon.

Amazon has developed a variety of robotic systems for use in their warehouses, including the Kiva robot, which transports shelves of goods to workers for order fulfilment. These robots are not humanoid in shape but they are designed to work alongside human workers in a collaborative manner, increasing efficiency and reducing the physical strain on human workers. Amazon already uses robot arms called ‘Robin’ and ‘Cardinal’ to re-direct boxed-up pre-delivery items around its warehouses and has also been testing its machine-learning ‘Sparrow’ robot arm that could handle 65 percent of its 100+ million diverse parcels. The Sparrow robot arm leverages the technologies of computer vision and artificial intelligence (AI) to help it detect, recognise, select, and handle a huge variety of different shapes and sizes of products prior to packaging, in the handling part of its business

Tesla, on the other hand, has developed robotic systems for use in their factories, including the TeslaBot, a humanoid robot that the company announced in 2021. The TeslaBot was intended to perform repetitive and dangerous tasks in factories, freeing up human workers to focus on more complex and creative work.

While Amazon and Tesla are not primarily focused on developing humanoid robots for a wide range of applications, they are both investing in developing robotic technology to increase efficiency and safety in their respective industries.

AI’s Game-Changing Abilities 

Although the game-changing success of generative AI chatbot technology such as that in ChatGPT has helped businesses and holds huge potential in combination with robotics, some think it’s time to put the brakes on and think about the consequences. For example, a recent open letter signed by more than 2600 influential people, including Elon Musk, called for 6-month moratorium on the training of AI systems more powerful than GPT-4.

Will A Robot Take Your Job? 

The potential impact of robots on employment is a hot topic for debate but while robots and automation can increase efficiency and productivity in certain industries, it is also true that they can replace some jobs that were previously done by humans. However, it’s important to note that robots are not a threat to all jobs, and that they are more likely to replace certain tasks within jobs rather than entire occupations. It’s also worth considering that robots and automation can create new jobs, both directly in the development, manufacturing, and maintenance of robots and their components, as well as indirectly in industries that benefit from increased efficiency and productivity.

In 2020, a World Economic Forum report estimated that by 2025, 85 million jobs may be displaced by a shift in labour division between humans and machines. The same report, however, predicted that 97 million new types of “jobs of the future” will open up as robots are used to take on the more mundane and repetitive tasks that they are more suited to than humans.

What Does This Mean For Your Business? 

Robotic automation in the workplace has been happening for many years and many companies recognise the value in automation that can perform repetitive tasks continuously and well, thereby bringing benefits like efficiency, safety, and 24-hour / 365 days a year working, reducing labour costs, and improving competitiveness.

As is usually the case in the tech world, it is the successful combination of different powerful technologies that brings about game-changing innovation, and the combination of AI, the power of which has been witnessed by all with ChatGPT, and robotics is exciting. Making robots that are ‘humanoid/android,’ however, feels like it’s one step closer to a science fiction future that’s been the stuff of films and dreams (nightmares?) for decades. In practical terms, a humanoid form is a design that is dextrous and the advantages in carrying out jobs is clear, although the worries about the advancement of AI could make the prospect of robots that are the most like humans to date a little creepy. That said, if the robots are affordable and can be shown to add real value in the workplace, there are likely to be many industries that could deploy them successfully. Although the current crop of robots is intended to augment the workforce, many human workers may be apprehensive that automation is coming for their jobs and may not yet be reassured that their introduction could create new types of jobs.